AI Apps Like Chat Gpt

AI Apps Like Chat Gpt — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

• 

  Rclone

  Rclone is an open source, multi threaded, command line computer program to manage or migrate content on cloud and other high latency storage. Its capabilities include sync, transfer, crypt, cache, union, compress and mount. The rclone website lists supported backends including S3 and Google Drive. Descriptions of rclone often carry the strapline "Rclone syncs your files to cloud storage". Those prior to 2020 include the alternative "Rsync for Cloud Storage". Rclone is well known for its rclone sync and rclone mount commands. It provides further management functions analogous to those ordinarily used for files on local disks, but which tolerate some intermittent and unreliable service. Rclone is commonly used with media servers such as Plex, Emby or Jellyfin to stream content direct from consumer file storage services. Official Ubuntu, Debian, Fedora, Gentoo, Arch, Brew, Chocolatey, and other package managers include rclone. == History == Nick Craig-Wood was inspired by rsync. Concerns about the noise and power costs arising from home computer servers prompted him to embrace cloud storage and he began developing rclone as open source software in 2012 under the name swiftsync. Rclone was promoted to stable version 1.00 in July 2014. In May 2017, Amazon Drive barred new users of rclone and other upload utilities, citing security concerns. Amazon Drive had been advertised as offering unlimited storage for £55 per year. Amazon's AWS S3 service continues to support new rclone users. The original rclone logo was updated in September 2018. In March 2020, Nick Craig-Wood resigned from Memset Ltd, a cloud hosting company he founded, to focus on open source software. Amazon's AWS April 2020 public sector blog explained how the Fred Hutch Cancer Research Center were using rclone in their Motuz tool to migrate very large biomedical research datasets in and out of AWS S3 object stores. In November 2020, rclone was updated to correct a weakness in the way it generated passwords. Passwords for encrypted remotes can be generated randomly by rclone or supplied by the user. In all versions of rclone from 1.49.0 to 1.53.2 the seed value for generated passwords was based on the number of seconds elapsed in the day, and therefore not truly random. CVE-2020-28924 recommended users upgrade to the latest version of rclone and check the passwords protecting their encrypted remotes. Release 1.55 of rclone in March 2021 included features sponsored by CERN and their CS3MESH4EOSC project. The work was EU funded to promote vendor-neutral application programming interfaces and protocols for synchronisation and sharing of academic data on cloud storage. == Backends and commands == Rclone supports the following services as backends. There are others, built on standard protocols such as WebDAV or S3, that work. WebDAV backends do not support rclone functionality dependent on server side checksum or modtime. Remotes are usually defined interactively from these backends, local disk, or memory (as S3), with rclone config. Rclone can further wrap those remotes with one or more of alias, chunk, compress, crypt or union, remotes. Once defined, the remotes are referenced by other rclone commands interchangeably with the local drive. Remote names are followed by a colon to distinguish them from local drives. For example, a remote example_remote containing a folder, or pseudofolder, myfolder is referred to within a command as a path example_remote:/myfolder. Rclone commands directly apply to remotes, or mount them for file access or streaming. With appropriate cache options the mount can be addressed as if a conventional, block level disk. Commands are provided to serve remotes over SFTP, HTTP, WebDAV, FTP and DLNA. Commands can have sub-commands and flags. Filters determine which files on a remote that rclone commands are applied to. rclone rc passes commands or new parameters to existing rclone sessions and has an experimental web browser interface. === Crypt remotes === Rclone's crypt implements encryption of files at rest in cloud storage. It layers an encrypted remote over a pre-existing, cloud or other remote. Crypt is commonly used to encrypt / decrypt media, for streaming, on consumer storage services such as Google Drive. Rclone's configuration file contains the crypt password. The password can be lightly obfuscated, or the whole rclone.conf file can be encrypted. Crypt can either encrypt file content and name, or additionally full paths. In the latter case there is a potential clash with encryption for cloud backends, such as Microsoft OneDrive, having limited path lengths. Crypt remotes do not encrypt object modification time or size. The encryption mechanism for content, name and path is available, for scrutiny, on the rclone website. Key derivation is with scrypt. === Example syntax (Linux) === These examples describe paths and file names but object keys behave similarly. To recursively copy files from directory remote_stuff, at the remote xmpl, to directory stuff in the home folder:- -v enables logging and -P, progress information. By default rclone checks the file integrity (hash) after copy; can retry each file up to three times if the operation is interrupted; uses up to four parallel transfer threads, and does not apply bandwidth throttling. Running the above command again copies any new or changed files at the remote to the local folder but, like default rsync behaviour, will not delete from the local directory, files which have been removed from the remote. To additionally delete files from the local folder which have been removed from the remote - more like the behaviour of rsync with a --delete flag:- And to delete files from the source after they have been transferred to the local directory - more like the behaviour of rsync with a --remove-source-file flag:- To mount the remote directory at a mountpoint in the pre-existing, empty stuff directory in the home directory (the ampersand at the end makes the mount command run as a background process):- Default rclone syntax can be modified. Alternative transfer, filter, conflict and backend specific flags are available. Performance choices include number of concurrent transfer threads; chunk size; bandwidth limit profiling, and cache aggression. == Academic evaluation == In 2018, University of Kentucky researchers published a conference paper comparing use of rclone and other command line, cloud data transfer agents for big data. The paper was published as a result of funding by the National Science Foundation. Later that year, University of Utah's Center for High Performance Computing examined the impact of rclone options on data transfer rates. == Rclone use at HPC research sites == Examples are University of Maryland, Iowa State University, Trinity College Dublin, NYU, BYU, Indiana University, CSC Finland, Utrecht University, University of Nebraska, University of Utah, North Carolina State University, Stony Brook, Tulane University, Washington State University, Georgia Tech, National Institutes of Health, Wharton, Yale, Harvard, Minnesota, Michigan State, Case Western Reserve University, University of South Dakota, Northern Arizona University, University of Pennsylvania, Stanford, University of Southern California, UC Santa Barbara, UC Irvine, UC Berkeley, and SURFnet. == Rclone and cybercrime == May 2020 reports stated rclone had been used by hackers to exploit Diebold Nixdorf ATMs with ProLock ransomware. The FBI issued a Flash Alert MI-000125-MW on May 4, 2020, in relation to the compromise. They issued a further, related alert 20200901–001 in September 2020. Attackers had exfiltrated / encrypted data from organisations involved in healthcare, construction, finance, and legal services. Multiple US government agencies, and industrial entities were affected. Researchers established the hackers spent about a month exploring the breached networks, using rclone to archive stolen data to cloud storage, before encrypting the target system. Reported targets included LaSalle County, and the city of Novi Sad. The FBI warned January 2021, in Private Industry Notification 20210106–001, of extortion activity using Egregor ransomware and rclone. Organisations worldwide had been threatened with public release of exfiltrated data. In some cases rclone had been disguised under the name svchost. Bookseller Barnes & Noble, US retailer Kmart, games developer Ubisoft and the Vancouver metro system have been reported as victims. An April 2021, cybersecurity investigation into SonicWall VPN zero-day vulnerability SNWLID-2021-0001 by FireEye's Mandiant team established attackers UNC2447 used rclone for reconnaissance and exfiltration of victims' files. Cybersecurity and Infrastructure Security Agency Analysis Report AR21-126A confirmed this use of rclone in FiveHands ransomware attacks. A June 2021, Microsoft Security Intelligence Twitter post identified use of rclone in BazaCall cyber attacks. The attackers sent emails e

  Read more →
• 

  Frameserver

  A frameserver is any program that acts as a media source in the process called frameserving, which transfers digital video data from one computer program to another without intermediate files. The program that receives the data – the frameclient – could be any type of video application. The process is controlled by the frameclient: the frameclient requests audio/video frames and the frameserver serves them. The client can request frames in any order, allowing it to pause or jump to an arbitrary frame, just as a media player does with a file on disk. The client is most commonly a media encoder, a non-linear editing system, or a media player. == Frameservers == AviSynth VirtualDub VapourSynth Debugmode FrameServer

  Read more →
• 

  Level-set method

  The Level-set method (LSM) is a conceptual framework for using level sets as a tool for numerical analysis of surfaces and shapes. LSM can perform numerical computations involving curves and surfaces on a fixed Cartesian grid without having to parameterize these objects. LSM makes it easier to perform computations on shapes with sharp corners and shapes that change topology (such as by splitting in two or developing holes). These characteristics make LSM effective for modeling objects that vary in time, such as an airbag inflating or a drop of oil floating in water. == Overview == The figure on the right illustrates several ideas about LSM. In the upper left corner is a bounded region with a well-behaved boundary. Below it, the red surface is the graph of a level set function φ {\displaystyle \varphi } determining this shape, and the flat blue region represents the X-Y plane. The boundary of the shape is then the zero-level set of φ {\displaystyle \varphi } , while the shape itself is the set of points in the plane for which φ {\displaystyle \varphi } is positive (interior of the shape) or zero (at the boundary). In the top row, the shape's topology changes as it is split in two. It is challenging to describe this transformation numerically by parameterizing the boundary of the shape and following its evolution. An algorithm can be used to detect the moment the shape splits in two and then construct parameterizations for the two newly obtained curves. On the bottom row, however, the plane at which the level set function is sampled is translated upwards, on which the shape's change in topology is described. It is less challenging to work with a shape through its level-set function rather than with itself directly, in which a method would need to consider all the possible deformations the shape might undergo. Thus, in two dimensions, the level-set method amounts to representing a closed curve Γ {\displaystyle \Gamma } (such as the shape boundary in our example) using an auxiliary function φ {\displaystyle \varphi } , called the level-set function. The curve Γ {\displaystyle \Gamma } is represented as the zero-level set of φ {\displaystyle \varphi } by Γ = { ( x , y ) ∣ φ ( x , y ) = 0 } , {\displaystyle \Gamma =\{(x,y)\mid \varphi (x,y)=0\},} and the level-set method manipulates Γ {\displaystyle \Gamma } implicitly through the function φ {\displaystyle \varphi } . This function φ {\displaystyle \varphi } is assumed to take positive values inside the region delimited by the curve Γ {\displaystyle \Gamma } and negative values outside. == The level-set equation == If the curve Γ {\displaystyle \Gamma } moves in the normal direction with a speed v {\displaystyle v} , then by chain rule and implicit differentiation, it can be determined that the level-set function φ {\displaystyle \varphi } satisfies the level-set equation ∂ φ ∂ t = v | ∇ φ | . {\displaystyle {\frac {\partial \varphi }{\partial t}}=v|\nabla \varphi |.} Here, | ⋅ | {\displaystyle |\cdot |} is the Euclidean norm (denoted customarily by single bars in partial differential equations), and t {\displaystyle t} is time. This is a partial differential equation, in particular a Hamilton–Jacobi equation, and can be solved numerically, for example, by using finite differences on a Cartesian grid. However, the numerical solution of the level set equation may require advanced techniques. Simple finite difference methods fail quickly. Upwinding methods such as the Godunov method are considered better; however, the level set method does not guarantee preservation of the volume and shape of the set level in an advection field that maintains shape and size, for example, a uniform or rotational velocity field. Instead, the shape of the level set may become distorted, and the level set may disappear over a few time steps. Therefore, high-order finite difference schemes, such as high-order essentially non-oscillatory (ENO) schemes, are often required, and even then, the feasibility of long-term simulations is questionable. More advanced methods have been developed to overcome this; for example, combinations of the leveling method with tracking marker particles suggested by the velocity field. == Example == Consider a unit circle in R 2 {\textstyle \mathbb {R} ^{2}} , shrinking in on itself at a constant rate, i.e. each point on the boundary of the circle moves along its inwards pointing normally at some fixed speed. The circle will shrink and eventually collapse down to a point. If an initial distance field is constructed (i.e. a function whose value is the signed Euclidean distance to the boundary, positive interior, negative exterior) on the initial circle, the normalized gradient of this field will be the circle normal. If the field has a constant value subtracted from it in time, the zero level (which was the initial boundary) of the new fields will also be circular and will similarly collapse to a point. This is due to this being effectively the temporal integration of the Eikonal equation with a fixed front velocity. == Applications == In mathematical modeling of combustion, LSM is used to describe the instantaneous flame surface, known as the G equation. Level-set data structures have been developed to facilitate the use of the level-set method in computer applications. Computational fluid dynamics Trajectory planning Optimization Image processing Computational biophysics Discrete complex dynamics (visualization of the parameter plane and the dynamic plane) == History == The level-set method was developed in 1979 by Alain Dervieux, and subsequently popularized by Stanley Osher and James Sethian. It has since become popular in many disciplines, such as image processing, computer graphics, computational geometry, optimization, computational fluid dynamics, and computational biology.

  Read more →
• 

  Comparison of vector graphics editors

  A number of vector graphics editors exist for various platforms. Potential users of these editors will make comparisons based on factors such as the availability for the user's platform, the software license, the feature set, the merits of the user interface (UI) and the focus of the program. Some programs are more suitable for artistic work while others are better for technical drawings. Another important factor is the application's support of various vector and bitmap image formats for import and export. The tables in this article compare general and technical information for a number of vector graphics editors. See the article on each editor for further information. This article is neither all-inclusive nor necessarily up-to-date. == Some editors in detail == Adobe Fireworks (formerly Macromedia Fireworks) is a vector editor with bitmap editing capabilities with its main purpose being the creation of graphics for Web and screen. Fireworks supports RGB color scheme and has no CMYK support. This means it is mostly used for screen design. The native Fireworks file format is editable PNG (FWPNG or PNG). Adobe Fireworks has a competitive price, but its features can seem limited in comparison with other products. It is easier to learn than other products and can produce complex vector artwork. The Fireworks editable PNG file format is not supported by other Adobe products. Fireworks can manage the PSD and AI file formats which enables it to be integrated with other Adobe apps. Fireworks can also open FWPNG/PNG, PSD, AI, EPS, JPG, GIF, BMP, TIFF file formats, and save/export to FWPNG/PNG, PSD, AI (v.8), FXG (v.2.0), JPG, GIF, PDF, SWF and some others. Some support for exporting to SVG is available via a free Export extension. On May 6, 2013, Adobe announced that Fireworks would be phased out. Adobe Flash (formerly a Macromedia product) has straightforward vector editing tools that make it easier for designers and illustrators to use. The most important of these tools are vector lines and fills with bitmap-like selectable areas, simple modification of curves via the "selection" or the control points/handles through "direct selection" tools. Flash uses Actionscript for OOP, and has full XML functionality through E4X support. Adobe FreeHand (formerly Macromedia Freehand and Aldus Freehand) is mainly used by professional graphic designers. The functionality of FreeHand includes the flexibility of the application in the wide design environment, catering to the output needs of both traditional image reproduction methods and to contemporary print and digital media with its page-layout capabilities and text attribute controls. Specific functions of FreeHand include a superior image-tracing operation for vector editing, page layout features within multiple-page documents, and embedding custom print-settings (such as variable halftone-screen specifications within a single graphic, etc.) to each document independent of auxiliary printer-drivers. User-operation is considered to be more suited for designers with an artistic background compared to designers with a technical background. When being marketed, FreeHand lacked the promotional backing, development and PR support in comparison to other similar products. FreeHand was transferred to the classic print group after Macromedia was purchased by Adobe in 2005. On May 16, 2007, Adobe announced that no further updates to Freehand would be developed but continues to sell FreeHand MX as a Macromedia product. FreeHand continues to run on Mac OS X Snow Leopard (using an Adobe fix) and on Windows 7. For macOS, Affinity Designer is able to open version 10 & MX Freehand files. Adobe Illustrator is a commonly used editor because of Adobe's market dominance, but is more expensive than other similar products. It is primarily developed consistently in line with other Adobe products and is best integrated with Adobe's Creative Suite packages. The ai file format is proprietary, but some vector editors can open and save in that format. Illustrator imports over two dozen formats, including PSD, PDF and SVG, and exports AI, PDF, SVG, SVGZ, GIF, JPG, PNG, WBMP, and SWF. However, the user must be aware of unchecking the "Preserve Illustrator Editing Capabilities" option if generating interoperable SVG files is desired. Affinity Designer by Serif Europe (the successor to their previous product, DrawPlus) is non-subscription-based software that is often described as an alternative to Adobe Illustrator. The application can open Portable Document Format (PDF), Adobe Photoshop, and Adobe Illustrator files, as well as export to those formats and to the Scalable Vector Graphics (SVG) and Encapsulated PostScript (EPS) formats. It also supports import from some Adobe Freehand files (specifically versions 10 & MX). Apache OpenOffice Draw is the vector graphics editor of the Apache OpenOffice open source office suite. It supports many import and export file formats and is available for multiple desktop operating systems. Boxy SVG is a chromium-based vector graphics editor for creating illustrations, as well as logos, icons, and other elements of graphic design. It is primarily focused on editing drawings in the SVG file format. The program is available as both a web app and a desktop application for Windows, macOS, ChromeOS, and Linux-based operating systems. Collabora Online Draw is the vector graphics editor of the Collabora Online open source office suite. It supports many import and export file formats and is accessible via any modern web browser, it also supports desktop editing features, Collabora Office is available for desktop and mobile operating systems, it is the enterprise ready version of LibreOffice. ConceptDraw PRO is a business diagramming tool and vector graphics editor available for both Windows and macOS. It supports multi-page documents, and includes an integrated presentation mode. ConceptDraw PRO supports imports and exports several formats, including Microsoft Visio and Microsoft PowerPoint. Corel Designer (originally Micrografx Designer) is one of the earliest vector-based graphics editors for the Microsoft Windows platform. The product is mainly used for the creation of engineering drawings and is shipped with extensive libraries for the needs of engineers. It is also flexible enough for most vector graphics design applications. CorelDRAW is an editor used in the graphic design, sign making and fashion design industries. CorelDRAW is capable of limited interoperation by reading file formats from Adobe Illustrator. CorelDRAW has over 50 import and export filters, on-screen and dialog box editing and the ability to create multi-page documents. It can also generate TrueType and Type 1 fonts, although refined typographic control is better suited to a more specific application. Some other features of CorelDRAW include the creation and execution of VBA macros, viewing of colour separations in print preview mode and integrated professional imposing options. Dia is a free and open-source diagramming and vector graphics editor available for Windows, Linux and other Unix-based computer operating systems. Dia has a modular design and several shape packages for flowcharting, network diagrams and circuit diagrams. Its design was inspired by Microsoft Visio, although it uses a Single Document Interface similar to other GNOME software (such as GIMP). DrawPlus, first built for the Windows platform in 1993, has matured into a full featured vector graphics editor for home and professional users. Also available as a feature-limited free 'starter edition': DrawPlus SE. DrawPlus developers, Serif Europe, have now ceased its development in order to focus on its successor, Affinity Designer. Edraw Max is a cross-platform diagram software and vector graphics editor available for Windows, Mac and Linux. It supports kinds of diagram types. It supports imports and exports SVG, PDF, HTML, Multiple page TIFF, Microsoft Visio and Microsoft PowerPoint. Embroidermodder is a free machine embroidery software tool that supports a variety of formats and allows the user to add custom modifications to their embroidery designs. Fatpaint is a free, light-weight, browser-based graphic design application with built-in vector drawing tools. It can be accessed through any browser with Flash 9 installed. Its integration with Zazzle makes it particularly suitable for people who want to create graphics for custom printed products such as T-shirts, mugs, iPhone cases, flyers and other promotional products. Figma is a collaborative web-based online vector graphics editor, used primarily for UX design and prototyping. GIMP, which works mainly with raster images, offers a limited set of features to create and record SVG files. It can also load and handle SVG files created with other software like Inkscape. Inkscape is a free and open-source vector editor with the primary native format being SVG. Inkscape is available for Linux, Windows, Mac OS X, and

  Read more →
• 

  AI agent

  In the context of generative artificial intelligence, AI agents (also referred to as compound AI systems or agentic AI) are a class of intelligent agents that can pursue goals, use tools, and take actions with varying degrees of autonomy. In practice, they usually operate within human-defined objectives, constraints, and available tools. == Overview == AI agents possess several key attributes, including goal-directed behavior, natural language interfaces, the capacity to use external tools, and the ability to perform multi-step tasks. Their control flow is frequently driven by large language models (LLMs). Agent systems may also include memory components, planning logic, tool interfaces, and orchestration software for coordinating agent components. AI agents do not have a standard definition. NIST describes agentic AI as an emerging area requiring standards for secure operation, interoperability, and reliable interaction with external systems. A common application of AI agents is task automation: for example, booking travel plans based on a user's prompted request. Companies such as Google, Microsoft and Amazon Web Services have offered platforms for deploying pre-built AI agents. Several protocols have been proposed for standardizing inter-agent communication, with examples including the Model Context Protocol, Gibberlink, and many others. Some of these protocols are also used for connecting agents to external applications. In December 2025, Linux Foundation announced the formation of the Agentic AI Foundation (AAIF), with the goal of ensuring agentic AI evolves transparently and collaboratively. == History == AI agents have been traced back to research from the 1990s, with Harvard professor Milind Tambe noting that the definition of an AI agent was not clear at the time. Researcher Andrew Ng has been credited with spreading the term "agentic" to a wider audience in 2024. == Training and testing == Researchers have attempted to build world models and reinforcement learning environments to train or evaluate AI agents. For example, video games such as Minecraft and No Man's Sky as well as replicas of company websites, have also been used for training such agents. == Autonomous capabilities == The Financial Times compared the autonomy of AI agents to the SAE classification of self-driving cars, likening most applications to level 2 or level 3, with some achieving level 4 in highly specialized circumstances, and level 5 being theoretical. == Cognitive architecture == The following are some internal design options for reasoning within an agent: Retrieval-augmented generation ReAct (Reason + Act) pattern is an iterative process in which an AI agent alternates between reasoning and taking actions, receives observations from the environment or external tools, and integrates these observations into subsequent reasoning steps. Reflexion, which uses an LLM to create feedback on the agent's plan of action and stores that feedback in a memory cache. A tool/agent registry, for organizing software functions or other agents that the agent can use. One-shot model querying, which queries the model once to create the plan of action. === Reference architecture === Ken Huang proposed an AI agent reference architecture, which consists of seven interconnected layers, with each layer building on the functionality of the layers beneath it: Layer 1: Foundation models - provide the core AI engines to power agent capabilities. Layer 2: Data operations - manage the complex data infrastructure required for AI agent operations, including Vector database, data loaders, RAG. Layer 3: Agent frameworks - sophisticated software and tools that simplify the development and management of the AI agents. Layer 4: Deployment and infrastructure - provide the robust technical foundation for running AI agents. Layer 5: Evaluation and observability - focus on assessing the safety and performance of AI agents. Layer 6: Security and compliance - a crucial protective framework ensuring AI agents operate safely, securely, and conform to regulatory boundaries. At this layer security and compliance features embedded into all the AI agent stack layers are integrated together. Layer 7: Agent ecosystem - represents the AI agents' interface with real-world applications and users. == Orchestration patterns == To execute complex tasks, autonomous agents are often integrated with other agents or specialized tools. These configurations, known as orchestration patterns or workflows, include the following: Prompt chaining: A sequence where the output of one step serves as the input for the next. Routing: The classification of an input to direct it to a specialized downstream task or tool. Parallelization: The simultaneous execution of multiple tasks. Sequential processing: A fixed, linear progression of tasks through a predefined pipeline. Planner-critic: An iterative pattern where one agent generates a proposal and another evaluates it to provide feedback for refinement. == Multimodal AI agents == In addition to large language models (LLMs), vision-language models (VLMs) and multimodal foundation models can be used as the basis for agents. In September 2024, Allen Institute for AI released an open-source vision-language model. Nvidia released a framework for developers to use VLMs, LLMs and retrieval-augmented generation for building AI agents that can analyze images and videos, including video search and video summarization. Microsoft released a multimodal agent model – trained on images, video, software user interface interactions, and robotics data – that the company claimed can manipulate software and robots. == Applications == As of April 2025, per the Associated Press, there are few real-world applications of AI agents. As of June 2025, per Fortune, many companies are primarily experimenting with AI agents. The Information divided AI agents into seven archetypes: business-task agents, for acting within enterprise software; conversational agents, which act as chatbots for customer support; research agents, for querying and analyzing information (such as OpenAI Deep Research); analytics agents, for analyzing data to create reports; software developer or coding agents (such as Cursor); domain-specific agents, which include specific subject matter knowledge; and web browser agents (such as OpenAI Operator). By mid-2025, AI agents have been used in video game development, gambling (including sports betting), cryptocurrency wallets (including cryptocurrency trading and meme coins) and social media. In August 2025, New York Magazine described software development as the most definitive use case of AI agents. Likewise, by October 2025, noting a decline in expectations, The Information noted AI coding agents and customer support as the primary use cases by businesses. In November 2025, The Wall Street Journal reported that few companies that deployed AI agents have received a return on investment. === Applications in government === Several government bodies in the United States and United Kingdom have deployed or announced the deployment of agents, at the local and national level. The city of Kyle, Texas deployed an AI agent from Salesforce in March 2025 for 311 customer service. In November 2025, the Internal Revenue Service stated that it would use Agentforce, AI agents from Salesforce, for the Office of Chief Counsel, Taxpayer Advocate Services and the Office of Appeals. That same month, Staffordshire Police announced that they would trial Agentforce agents for handling non-emergency 101 calls in the United Kingdom starting in 2026. In December 2025, the Department of Neighborhoods in Detroit, Michigan, in partnership with a local business, deployed a pilot project in two Detroit districts for an AI agent to be used for customer service calls. In February 2025, Thomas Shedd, the director of the Technology Transformation Services, proposed using AI coding agents across the United States federal government. A recruiter for the Department of Government Efficiency proposed in April 2025 to use AI agents to automate the work of about 70,000 United States federal government employees, as part of a startup with funding from OpenAI and a partnership agreement with Palantir. This proposal was criticized by experts for its impracticality, if not impossibility, and the lack of corresponding widespread adoption by businesses. In December 2025, the Food and Drug Administration announced that it would offer "agentic AI capabilities" to its staff for "meeting management, pre-market reviews, review validation, post-market surveillance, inspections and compliance and administrative functions." That same month, the United States Department of Defense launched GenAI.mil, an internal platform for American military personnel to use generative AI-based applications based on Google Gemini, including "intelligent agentic workflows". Defense Secretary Pete Hegseth listed applications such as "[conducting] deep r

  Read more →
• 

  Pixel shift

  Pixel shift is a method in digital cameras for producing a super-resolution image. The method works by taking several images, after each such capture moving ("shifting") the sensor to a new position. In digital colour cameras that employ pixel shift, this avoids a major limitation inherent in using Bayer pattern for obtaining colour, and instead produces an image with increased colour resolution and, assuming a static subject or additional computational steps, an image free of colour moiré. Taking this idea further, sub-pixel shifting may increase the resolution of the final image beyond that suggested by the specified resolution of the image sensor. Additionally, assuming that the various individual captures are taken at the same sensitivity, the final combined image will have less image noise than a single capture. This can be thought of as an averaging effect (for instance, in a pixel shift image composed of four individual frames with a classic Bayer pattern, every pixel in the final colour image is based on two measurements of the green channel). == List of cameras implementing pixel shift == All of the following cameras are fabricated with one imaging sensor, thus any kind of pixel shift requires a movement of the whole sensor. === Canon === Canon R5: Contains a 45 Mpixel sensor. The High-Resolution Mode shifts the sensor by one pixel to obtain a sequence of nine images that are merged into a 400 Mpixel image. === Fujifilm === Fujifilm GFX50S II: contains a 51 Mpixel sensor. The Pixel Shift Multi-Shot mode shifts the imaging sensor by 0.5-pixel movements to obtain a sequence of 16 images that are subsequently merged into a 200 Mpixel image. Fujifilm GFX100, Fujifilm GFX100 II: contains a 102 Mpixel sensor. A sequence of 16 pixel shifted images are merged into a 400 Mpixel image. Fujifilm GFX100S, Fujifilm GFX100S II: contains a 102 Mpixel sensor. A sequence of 16 pixel shifted images are merged into a 400 Mpixel image Fujifilm GFX100IR: contains a 102 Mpixel sensor. A sequence of 16 pixel shifted images are merged into a 400 Mpixel image Fujifilm X-H2: contains a 40 Mpixel sensor. A sequence of 20 shifted images are merged into a 160 Mpixel image. Fujifilm X-T5: contains a 40 Mpixel sensor. A sequence of 20 shifted images are merged into a 160 Mpixel image. === Nikon === Nikon Z8: contains a 47.5 Mpixel sensor. The High Res shot mode shifts the imaging sensor by 0.5-pixel movements to obtain a sequence of up to 32 images that can be merged in Nikon's NX studio software. Nikon Zf: contains a 24 Mpixel sensor. The High Res shot mode shifts the imaging sensor by 0.5-pixel movements to obtain a sequence of up to 32 images that can be merged in Nikon's NX studio software. === Olympus === Olympus OM-D E-M1 Mark II: contains a 20.4 Mpixel sensor. The High Res shot mode produces a 50 Mpixel image. Olympus OM-D E-M5 Mark II: contains a 16 Mpixel sensor. The High Res shot mode shifts the imaging sensor by 0.5-pixel movements to obtain a sequence of 8 images that are subsequently merged into a 40 Mpixel image. Olympus OM-D E-M5 Mark III: contains a 20.4 Mpixel sensor. The High Res shot mode shifts the imaging sensor by 0.5-pixel movements to obtain a sequence of 8 images that are subsequently merged into a 50 Mpixel image. Olympus OM-D E-M1X: contains a 20.4 Mpixel sensor. The camera sports two pixel shift mode: (a) the 80Mp Tripod mode produces an 80 Mpixel image, (b) the Handheld High Res shot mode produces a 50 Mpixel image. Olympus PEN-F: contains a 20.4 Mpixel sensor. The High Res Shot mode takes multiple images, continually shifting the position of the sensor in sub-pixel increments. Combining these images results in either a 50MP JPEG or an 80MP Raw file. ==== OM System ==== OM System OM-1: contains a 20MPix sensor. The High Res Shot mode takes multiple images, and it can be used handheld or on a tripod. Handheld it will internally produce 50 Mpix files and 80 Mpix when mounted on a tripod. OM System OM-5: contains a 20MPix sensor. The High Res Shot mode takes multiple images, and it can be used handheld or on a tripod. Handheld it will internally produce 50 Mpix files and 80 Mpix when mounted on a tripod. === Panasonic === Panasonic Lumix DC-G9: contains a 20.3 Mpixel sensor. The High Resolution Mode takes a sequence of 8 shots in quick succession between which the sensor is shifted by 0.5 pixel for each image. These are subsequently merged into an 80 Mpixel image. Panasonic Lumix DC-S1: contains a 24.2 Mpixel sensor. The High Resolution Mode takes a sequence of shots in quick succession between which the sensor is shifted by a small amount. These are subsequently merged into a 96 Mpixel image. Panasonic Lumix DC-S1R: contains a 47.3 Mpixel sensor. The High Resolution Mode shifts the imaging sensor by a small increments to obtain a sequence of 8 images that are subsequently merged into a 187 Mpixel image. Panasonic Lumix DC-S1H Panasonic Lumix DC-S5 === Pentax === Pentax K-70: contains a 24.3 Mpixel sensor. The pixel shift mode takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into an image sporting 'all color data in each pixel to deliver super-high-resolution images'. Pentax KP: contains a 24.3 Mpixel sensor. The pixel shift mode takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into an image sporting 'high-resolution images with more accurate colours and much finer details'. Pentax K-3 II: contains a 24.3 Mpixel sensor. The pixel shift mode takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into an image sporting 'super-high-resolution images with far more truthful color reproduction and much finer details'. Pentax K-3 III: contains a 25.7 Mpixel sensor. The pixel shift mode takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into an image sporting 'a cancelling out of the Bayer pattern and removal of the need for sharpness-sapping demosaicing'. Pentax K-1: contains a 36.4 Mpixel sensor. The pixel shift mode takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into an image sporting 'improved detail and colour resolution'. Pentax K-1 II: contains a 36.4 Mpixel sensor. The camera sports two pixel shift mode: (a) a series of 4 tripod-stabilised images shifted by 1 pixel each are subsequently combined into a 47.3 Mpixel image, (b) a series of images taken in handheld mode are combined into a 47.3 Mpixel image that is, within limits, able to cope even with moving subjects. === Sony === Sony a6600: contains a 24.3 Mpixel sensor. The pixel shift mode takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into an image sporting 'all color data in each pixel to deliver super-high-resolution images'. Sony α7R III: contains a 42.4 Mpixel sensor. The pixel shift mode takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into a 42.4 Mpixel image with improved tonal resolution. Sony α7R IV: contains a 61 Mpixel sensor. The camera has two pixel shift modes, (a) the first takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into a 61 Mpixel image with improved tonal resolution, (b) the other takes a sequence of 16 shots between which the sensor is shifted by 0.5 pixel. These are subsequently merged into a 240 Mpixel image with both enhanced detail and improved tonal resolution. Sony α1: contains a 50 Mpixel sensor. The camera has two pixel shift modes, (a) the first takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into a 50 Mpixel image with improved tonal resolution, (b) the other takes a sequence of 16 shots between which the sensor is shifted by 0.5 pixel. These are subsequently merged into a 200 Mpixel image with both enhanced detail and improved tonal resolution. === Hasselblad === Hasselblad H3DII: the model H3DII-39 sports a 39 Mpixel sensor, the model H3DII-50 a 50 Mpixel sensor. Both enable a pixel shift mode which takes a sequence of 4 shots between which the sensor is shifted by 1 pixel. These are subsequently merged into a single image. Hasselblad H4D series: the model H4D-200MS contains a 50 Mpixel sensor. The sensor sports 3 different pixel shift modes which take (a) a sequence of 6 shots taken at slight offsets, (b) a sequence of 4 shots between which the sensor is shifted by 1 pixel, (c) a sequence of 4 shots between which the sensor is shifted by 0.5 pixels. Images obtained by all three modes are subsequently merged into 200 Mpixel images. Hasselblad H5D series: both models H5D-50c MS and H5D-200c MS contain a 50 Mpixel sensor. This sensor sports 2 different pixel shift modes which take (a) a sequence of 6 shots with full and half pixel moveme

  Read more →
• 

  Plotting algorithms for the Mandelbrot set

  There are many programs and algorithms used to plot the Mandelbrot set and other fractals, some of which are described in fractal-generating software. These programs use a variety of algorithms to determine the color of individual pixels efficiently. == Escape time algorithm == The simplest algorithm for generating a representation of the Mandelbrot set is known as the "escape time" algorithm. A repeating calculation is performed for each x, y point in the plot area and based on the behavior of that calculation, a color is chosen for that pixel. === Unoptimized naïve escape time algorithm === In both the unoptimized and optimized escape time algorithms, the x and y locations of each point are used as starting values in a repeating, or iterating calculation (described in detail below). The result of each iteration is used as the starting values for the next. The values are checked during each iteration to see whether they have reached a critical "escape" condition, or "bailout". If that condition is reached, the calculation is stopped, the pixel is drawn, and the next x, y point is examined. For some starting values, escape occurs quickly, after only a small number of iterations. For starting values very close to but not in the set, it may take hundreds or thousands of iterations to escape. For values within the Mandelbrot set, escape will never occur. The programmer or user must choose how many iterations–or how much "depth"–they wish to examine. The higher the maximal number of iterations, the more detail and subtlety emerge in the final image, but the longer time it will take to calculate the fractal image. Escape conditions can be simple or complex. Because no complex number with a real or imaginary part greater than 2 can be part of the set, a common bailout is to escape when either coefficient exceeds 2. A more computationally complex method that detects escapes sooner, is to compute distance from the origin using the Pythagorean theorem, i.e., to determine the absolute value, or modulus, of the complex number. If this value exceeds 2, or equivalently, when the sum of the squares of the real and imaginary parts exceed 4, the point has reached escape. More computationally intensive rendering variations include the Buddhabrot method, which finds escaping points and plots their iterated coordinates. The color of each point represents how quickly the values reached the escape point. Often black is used to show values that fail to escape before the iteration limit, and gradually brighter colors are used for points that escape. This gives a visual representation of how many cycles were required before reaching the escape condition. To render such an image, the region of the complex plane we are considering is subdivided into a certain number of pixels. To color any such pixel, let c {\displaystyle c} be the midpoint of that pixel. We now iterate the critical point 0 under P c {\displaystyle P_{c}} , checking at each step whether the orbit point has modulus larger than 2. When this is the case, we know that c {\displaystyle c} does not belong to the Mandelbrot set, and we color our pixel according to the number of iterations used to find out. Otherwise, we keep iterating up to a fixed number of steps, after which we decide that our parameter is "probably" in the Mandelbrot set, or at least very close to it, and color the pixel black. In pseudocode, this algorithm would look as follows. The algorithm does not use complex numbers and manually simulates complex-number operations using two real numbers, for those who do not have a complex data type. The program may be simplified if the programming language includes complex-data-type operations. for each pixel (Px, Py) on the screen do x0 := scaled x coordinate of pixel (scaled to lie in the Mandelbrot X scale (-2.00, 0.47)) y0 := scaled y coordinate of pixel (scaled to lie in the Mandelbrot Y scale (-1.12, 1.12)) x := 0.0 y := 0.0 iteration := 0 max_iteration := 1000 while (xx + yy ≤ 22 AND iteration < max_iteration) do xtemp := xx - yy + x0 y := 2xy + y0 x := xtemp iteration := iteration + 1 color := palette[iteration] plot(Px, Py, color) Here, relating the pseudocode to c {\displaystyle c} , z {\displaystyle z} and P c {\displaystyle P_{c}} : z = x + i y {\displaystyle z=x+iy\ } z 2 = x 2 + 2 i x y {\displaystyle z^{2}=x^{2}+2ixy} - y 2 {\displaystyle y^{2}\ } c = x 0 + i y 0 {\displaystyle c=x_{0}+iy_{0}\ } and so, as can be seen in the pseudocode in the computation of x and y: x = R e ⁡ ( z 2 + c ) = x 2 − y 2 + x 0 {\displaystyle x=\mathop {\mathrm {Re} } (z^{2}+c)=x^{2}-y^{2}+x_{0}} and y = I m ⁡ ( z 2 + c ) = 2 x y + y 0 . {\displaystyle y=\mathop {\mathrm {Im} } (z^{2}+c)=2xy+y_{0}.\ } To get colorful images of the set, the assignment of a color to each value of the number of executed iterations can be made using one of a variety of functions (linear, exponential, etc.). One practical way, without slowing down calculations, is to use the number of executed iterations as an entry to a palette initialized at startup. If the color table has, for instance, 500 entries, then the color selection is n mod 500, where n is the number of iterations. === Optimized escape time algorithms === The code in the previous section uses an unoptimized inner while loop for clarity. In the unoptimized version, one must perform five multiplications per iteration. To reduce the number of multiplications the following code for the inner while loop may be used instead: x2:= 0 y2:= 0 w:= 0 while (x2 + y2 ≤ 4 and iteration < max_iteration) do x:= x2 - y2 + x0 y:= w - x2 - y2 + y0 x2:= x x y2:= y y w:= (x + y) (x + y) iteration:= iteration + 1 The above code works via some algebraic simplification of the complex multiplication: ( i y + x ) 2 = − y 2 + 2 i y x + x 2 = x 2 − y 2 + 2 i y x {\displaystyle {\begin{aligned}(iy+x)^{2}&=-y^{2}+2iyx+x^{2}\\&=x^{2}-y^{2}+2iyx\end{aligned}}} Using the above identity, the number of multiplications can be reduced to three instead of five. The above inner while loop can be further optimized by expanding w to w = x 2 + 2 x y + y 2 {\displaystyle w=x^{2}+2xy+y^{2}} Substituting w into y = w − x 2 − y 2 + y 0 {\displaystyle y=w-x^{2}-y^{2}+y_{0}} yields y = 2 x y + y 0 {\displaystyle y=2xy+y_{0}} and hence calculating w is no longer needed. The further optimized pseudocode for the above is: x:= 0 y:= 0 x2:= 0 y2:= 0 while (x2 + y2 ≤ 4 and iteration < max_iteration) do x2:= x x y2:= y y y:= 2 x y + y0 x:= x2 - y2 + x0 iteration:= iteration + 1 Note that in the above pseudocode, 2 x y {\displaystyle 2xy} seems to increase the number of multiplications by 1, but since 2 is the multiplier the code can be optimized via ( x + x ) y {\displaystyle (x+x)y} . == Coloring algorithms == In addition to plotting the set, a variety of algorithms have been developed to efficiently color the set in an aesthetically pleasing way show structures of the data (scientific visualisation) === Histogram coloring === A more complex coloring method involves using a histogram which pairs each pixel with said pixel's maximum iteration count before escape/bailout. This method will equally distribute colors to the same overall area, and, importantly, is independent of the maximum number of iterations chosen. This algorithm has four passes. The first pass involves calculating the iteration counts associated with each pixel (but without any pixels being plotted). These are stored in an array IterationCounts[x][y], where x and y are the x and y coordinates of said pixel on the screen respectively. The first step of the second pass is to create an array NumIterationsPerPixel[n], where the array size n is the maximum iteration count. Next, one must iterate over the array of pixel-iteration count pairs IterationCounts[x][y], and retrieve each pixel's saved iteration count, i, via e.g. i = IterationCounts[x][y]. After each pixel's iteration count i is retrieved, it is necessary to index the NumIterationsPerPixel array at i and increment the indexed value (which is initially zero) -- e.g. NumIterationsPerPixel[i] = NumIterationsPerPixel[i] + 1. for (x = 0; x < width; x++) do for (y = 0; y < height; y++) do i:= IterationCounts[x][y] NumIterationsPerPixel[i]++ The third pass iterates through the NumIterationsPerPixel array and adds up all the stored values, saving them in total. The array index represents the number of pixels that reached that iteration count before bailout. total: = 0 for (i = 0; i < max_iterations; i++) do total += NumIterationsPerPixel[i] After this, the fourth pass begins and all the values in the IterationCounts array are indexed, and, for each iteration count i, associated with each pixel, the count is added to a global sum of all the iteration counts from 1 to i in the NumIterationsPerPixel array . This value is then normalized by dividing the sum by the total value computed earlier. hue[][]:= 0.0 for (x = 0; x < width; x++) do for (y = 0; y < height; y++) do iteration:= Iteration

  Read more →
• 

  Speech recognition

  Speech recognition (automatic speech recognition (ASR), computer speech recognition, or speech-to-text (STT)) is a sub-field of computational linguistics concerned with methods and technologies that translate spoken language into text or other interpretable forms. Speech recognition applications include voice user interfaces, where the user speaks to a device, which "listens" and processes the audio. Common voice applications include interpreting commands for calling, call routing, home automation, and aircraft control. These applications are called direct voice input. Productivity applications include searching audio recordings, creating transcripts, and dictation. Speech recognition can be used to analyse speaker characteristics, such as identifying native language using pronunciation assessment. Voice recognition (speaker identification) refers to identifying the speaker, rather than speech contents. Recognizing the speaker can simplify the task of translating speech in systems trained on a specific person's voice. It can also be used to authenticate the speaker as part of a security process. == History == Applications for speech recognition developed over many decades, with progress accelerated due to advances in deep learning and the use of big data. These advances are reflected in an increase in academic papers, and greater system adoption. Key areas of growth include vocabulary size, more accurate recognition for unfamiliar speakers (speaker independence), and faster processing speed. === Pre-1970 === 1952 – Bell Labs researchers, Stephen Balashek, R. Biddulph, and K. H. Davis, built Audrey for single-speaker digit recognition. Their system located the formants in the power spectrum of each utterance. 1960 – Gunnar Fant developed and published the source–filter model of speech production. 1962 – IBM's 16-word "Shoebox" machine's speech recognition debuted at the 1962 World's Fair. 1966 – Linear predictive coding, a speech coding method, was proposed by Fumitada Itakura of Nagoya University and Shuzo Saito of Nippon Telegraph and Telephone. 1969 – Funding at Bell Labs came to a halt for several years after the company's head engineer, John R. Pierce, wrote an open letter criticizing speech recognition research. This defunding lasted until Pierce retired and James L. Flanagan took over. Raj Reddy was the first person to work on continuous speech recognition, as a graduate student at Stanford University in the late 1960s. Previous systems required users to pause after each word. Reddy's system issued spoken commands for playing chess. Around this time, Soviet researchers invented the dynamic time warping (DTW) algorithm and used it to create a recognizer capable of operating on a 200-word vocabulary. DTW processed speech by dividing it into short frames (e.g. 10 ms segments) and treating each frame as a unit. Speaker independence, however, remained unsolved. === 1970–1990 === 1971 – DARPA funded a five-year speech recognition research project, Speech Understanding Research, seeking a minimum vocabulary size of 1,000 words. The project considered speech understanding a key to achieving progress in speech recognition, which was later disproved. BBN, IBM, Carnegie Mellon (CMU), and Stanford Research Institute participated. 1972 – The IEEE Acoustics, Speech, and Signal Processing group held a conference in Newton, Massachusetts. 1976 – The first ICASSP was held in Philadelphia, which became a major venue for publishing on speech recognition. During the late 1960s, Leonard Baum developed the mathematics of Markov chains at the Institute for Defense Analysis. A decade later, at CMU, Raj Reddy's students James Baker and Janet M. Baker began using the hidden Markov model (HMM) for speech recognition. James Baker had learned about HMMs while at the Institute for Defense Analysis. HMMs enabled researchers to combine sources of knowledge, such as acoustics, language, and syntax, in a unified probabilistic model. By the mid-1980s, Fred Jelinek's team at IBM created a voice-activated typewriter called Tangora, which could handle a 20,000-word vocabulary. Jelinek's statistical approach placed less emphasis on emulating human brain processes in favor of statistical modelling. (Jelinek's group independently discovered the application of HMMs to speech.) This was controversial among linguists since HMMs are too simplistic to account for many features of human languages. However, the HMM proved to be a highly useful way for modelling speech and replaced dynamic time warping as the dominant speech recognition algorithm in the 1980s. 1982 – Dragon Systems, founded by James and Janet M. Baker, was one of IBM's few competitors. === Practical speech recognition === The 1980s also saw the introduction of the n-gram language model. 1987 – The back-off model enabled language models to use multiple-length n-grams, and CSELT used HMM to recognize languages (in software and hardware, e.g. RIPAC). At the end of the DARPA program in 1976, the best computer available to researchers was the PDP-10 with 4 MB of RAM. It could take up to 100 minutes to decode 30 seconds of speech. Practical products included: 1984 – the Apricot Portable was released with up to 4096 words support, of which only 64 could be held in RAM at a time. 1987 – a recognizer from Kurzweil Applied Intelligence 1990 – Dragon Dictate, a consumer product released in 1990. AT&T deployed the Voice Recognition Call Processing service in 1992 to route telephone calls without a human operator. The technology was developed by Lawrence Rabiner and others at Bell Labs. By the early 1990s, the vocabulary of the typical commercial speech recognition system had exceeded the average human vocabulary. Reddy's former student, Xuedong Huang, developed the Sphinx-II system at CMU. Sphinx-II was the first to do speaker-independent, large vocabulary, continuous speech recognition, and it won DARPA's 1992 evaluation. Handling continuous speech with a large vocabulary was a major milestone. Huang later founded the speech recognition group at Microsoft in 1993. Reddy's student Kai-Fu Lee joined Apple, where, in 1992, he helped develop the Casper speech interface prototype. Lernout & Hauspie, a Belgium-based speech recognition company, acquired other companies, including Kurzweil Applied Intelligence in 1997 and Dragon Systems in 2000. L&H was used in Windows XP. L&H was an industry leader until an accounting scandal destroyed it in 2001. L&H speech technology was bought by ScanSoft, which became Nuance in 2005. Apple licensed Nuance software for its digital assistant Siri. ==== 2000s ==== In the 2000s, DARPA sponsored two speech recognition programs: Effective Affordable Reusable Speech-to-Text (EARS) in 2002, followed by Global Autonomous Language Exploitation (GALE) in 2005. Four teams participated in EARS: IBM; a team led by BBN with LIMSI and the University of Pittsburgh; Cambridge University; and a team composed of ICSI, SRI, and the University of Washington. EARS funded the collection of the Switchboard telephone speech corpus, which contained 260 hours of recorded conversations from over 500 speakers. The GALE program focused on Arabic and Mandarin broadcast news. Google's first effort at speech recognition came in 2007 after recruiting Nuance researchers. Its first product, GOOG-411, was a telephone-based directory service. Since at least 2006, the U.S. National Security Agency has employed keyword spotting, allowing analysts to index large volumes of recorded conversations and identify speech containing "interesting" keywords. Other government research programs focused on intelligence applications, such as DARPA's EARS program and IARPA's Babel program. In the early 2000s, speech recognition was dominated by hidden Markov models combined with feed-forward artificial neural networks (ANN). Later, speech recognition was taken over by long short-term memory (LSTM), a recurrent neural network (RNN) published by Sepp Hochreiter & Jürgen Schmidhuber in 1997. LSTM RNNs avoid the vanishing gradient problem and can learn "Very Deep Learning" tasks that require memories of events that happened thousands of discrete time steps earlier, which is important for speech. Around 2007, LSTMs trained with Connectionist Temporal Classification (CTC) began to outperform. In 2015, Google reported a 49 percent error‑rate reduction in its speech recognition via CTC‑trained LSTM. Transformers, a type of neural network based solely on attention, were adopted in computer vision and language modelling, and then to speech recognition. Deep feed-forward (non-recurrent) networks for acoustic modelling were introduced in 2009 by Geoffrey Hinton and his students at the University of Toronto, and by Li Deng and colleagues at Microsoft Research. In contrast to the prioer incremental improvements, deep learning decreased error rates by 30%. Both shallow and deep forms (e.g., recurrent nets) of ANNs had been explored since the 1980s. Howev

  Read more →
• 

  Meta-Labeling

  Meta-labeling, also known as corrective AI, is a machine learning (ML) technique utilized in quantitative finance to enhance the performance of investment and trading strategies, developed in 2017 by Marcos López de Prado at Guggenheim Partners and Cornell University. The core idea is to separate the decision of trade direction (side) from the decision of trade sizing, addressing the inefficiencies of simultaneously learning both side and size predictions. The side decision involves forecasting market movements (long, short, neutral), while the size decision focuses on risk management and profitability. It serves as a secondary decision-making layer that evaluates the signals generated by a primary predictive model. By assessing the confidence and likely profitability of those signals, meta-labeling allows investors and algorithms to dynamically size positions and suppress false positives. == Motivation == Meta-labeling is designed to improve precision without sacrificing recall. As noted by López de Prado, attempting to model both the direction and the magnitude of a trade using a single algorithm can result in poor generalization. By separating these tasks, meta-labeling enables greater flexibility and robustness: Enhances control over capital allocation. Reduces overfitting by limiting model complexity. Allows the use of interpretability tools and tailored thresholds to manage risk. Enables dynamic trade suppression in unfavorable regimes. == Applications == Meta-labeling has been applied in a variety of financial ML contexts, including: Algorithmic trading: Filtering and sizing trades to reduce false positives. Portfolio optimization: Scaling exposure across multiple signals with differing confidence levels. Risk management: Dynamically disabling strategies in adverse market conditions. Model validation: Interpreting when and why a model may be underperforming due to regime shifts. == General architecture == Meta-labeling decouples two core components of systematic trading strategies: directional prediction and position sizing. The process involves training a primary model to generate trade signals (e.g., buy, sell, or hold) and then training a secondary model to determine whether each signal is likely to lead to a profitable trade. The second model outputs a probability that is interpreted as the confidence in the forecast, which can be used to adjust the position size or to filter out unreliable trades. Meta-labeling is typically implemented as a three-stage process: Primary model (M1): Predicts the direction or label of a financial outcome using features such as market prices, returns, or volatility indicators. A typical output is directional, e.g., Y ∈ {−1,0,1}, representing short, neutral, or long positions. Secondary model (M2): A binary classifier trained to predict whether the primary model's prediction will be profitable. The target variable is a binary meta-label F ∈ { 0 , 1 } {\displaystyle F\in \{0,1\}} . Inputs can include features used in the primary model, performance diagnostics, or market regime data. Position sizing algorithm (M3): Translates the output probability of the secondary model into a position size. Higher confidence scores result in larger allocations, while lower confidence leads to reduced or zero exposure. === Stage 1: Forecasting side === Primary model architecture Figure 1 Figure 1 presents the architecture of a primary model. It focuses on forecasting the side of the trade. Following the example, this model (M1) takes in input data – such as open-high-low-close data and determines the side of the position to take: a negative number is a short position, and positive number is a long position, the range is set between −1 and 1 (the closer it is to −1 or 1, the stronger the models conviction is). When training the model, the labels are −1 and 1, based on the direction of forward returns for some predefined investment horizon. The researcher may decide to apply a recall check (τ: "Tau") by setting a minimum threshold that the initial output needs to be to qualify of a short or long position (if the threshold is not met, no side forecast is predicted, leading to closing of any open positions), this leads to the primary model output which is one of three possible side forecasts: −1, 0, or 1. The primary model also generates evaluation data which can be used by the secondary model, to improve performance of size forecasts. Some examples of evaluation data include rolling accuracy, F1, recall, precision, and AUC scores. === Stage 2: Filtering out false positives === General meta-labeling architecture Figure 2 Next comes the phase of filtering out false positives, by applying a secondary machine learning model (M2), which is a binary classifier trained to determine if the trade will be profitable or not. The model takes as input four general groupings of data: General input data which is predictive of a false positive. For example the last 30 days rolling volatility of the underlying asset. Evaluation data. Market state and regime data, one may find that macro economic data or clustering the market into regimes may help as specific trading strategies are known to perform better in particular regimes. Example: momentum based strategies perform best in periods with low volatility and strong directional moves. Primary models initial input which is a value between −1 and 1. This highlights the strength of the primary models conviction. The output of the model is a value between −1 and 1 (if using a Tanh function) which will indicate the strength of the conviction that a short or long position is profitable, or it could simply be between 0 and 1 (using a sigmoid function) if one only wanted to know if it made money or not. This output allows filtering out trades that are likely to lead to losses. One could stop at this point or use the outputs of the secondary model as inputs to a position sizing algorithm (M3) which could further enhance strategy performance metrics by translating the output probability of the secondary model into a position size. Higher confidence scores result in larger allocations, while lower confidence leads to reduced or zero exposure. === Stage 3: Optimizing position sizes === ==== Position sizing methods (M3) ==== Various algorithms have been proposed for transforming predicted probabilities into trade sizes: All-or-nothing: Allocate 100% of capital if the probability exceeds a predefined threshold (e.g., 0.5); otherwise, do not trade. Model confidence: Use the probability score directly as the fraction of capital allocated. Linear scaling: Rescale the model's probabilities using min-max normalization based on the training data. Normal CDF (NCDF): Use a normal cumulative distribution function applied to a z-statistic derived from the predicted probability. Empirical CDF (ECDF): Rank probabilities based on their percentile in the training data to ensure relative allocation. Sigmoid Optimal Position Sizing (SOPS): Applies a smooth non-linear sigmoid transformation optimized to maximize risk-adjusted returns (Sharpe ratio). ==== Model calibration ==== Each machine learning algorithm used in meta-labeling tends to produce outputs with different characteristic distributions; for example, some are approximately normally distributed, whereas others exhibit a pronounced U-shape, concentrating probabilities near the extremes. Due to these varying distributions, simply summing the outputs of different models can inadvertently lead to uneven weighting of signals, biasing trade decisions. To address this, model calibration techniques are essential to adjust the predicted probabilities towards frequentist probabilities, ensuring that model outputs reflect true likelihoods more accurately. Two common calibration techniques are: Platt scaling (Sigmoid scaling): Suitable for correcting S-shaped calibration plots typically produced by models such as support vector machines (SVMs). Isotonic regression: Fits a non-decreasing step function to probabilities and is effective particularly with larger datasets, though it can sometimes lead to overfitting. Transforming predictions to frequentist probabilities is crucial as it provides probabilistic outputs that are directly interpretable as the actual likelihood of an event occurring. Such calibration significantly enhances the effectiveness of fixed position sizing methods, reducing maximum drawdowns and increasing risk-adjusted returns. However, calibration has less impact on position sizing methods that directly estimate parameters from the training data, such as ECDF and SOPS, suggesting that calibration is a critical step mainly for fixed methods that rely heavily on raw model outputs. =

  Read more →
• 

  Parasolid

  Parasolid is a geometric modeling kernel originally developed by Shape Data Limited, now owned and developed by Siemens Digital Industries Software. It can be licensed by other companies for use in their 3D computer graphics software products. Parasolid's abilities include model creation and editing utilities such as Boolean modeling operators, feature modeling support, advanced surfacing, thickening and hollowing, blending and filleting, and sheet modeling. It also incorporates modeling with mesh surfaces and lattices. Parasolid also includes tools for direct model editing, including tapering, offsetting, geometry replacement and removing feature details with automated regeneration of surrounding data. Parasolid also provides wide-ranging graphical and rendering support, including hidden-line, wireframe and drafting, tessellation, and model data inquiries. To use Parasolid effectively, software developers need knowledge of CAD in general, computational geometry, and topology. Parasolid is available for Windows (32-bit, 64-bit and AArch64), Linux (64-bit and AArch64), macOS (Apple silicon and Intel), iOS, and Android. == Parasolid XT format == Parasolid parts are normally saved in XT format, which usually has the file extension .X_T. The format is documented and open. There is also a binary version of the format, usually with an .X_B extension, which is somewhat more compact. Both .X_T and .X_B are used for parts files. == Applications == It is used in many computer-aided design (CAD), computer-aided manufacturing (CAM), computer-aided engineering (CAE), product visualization, and CAD data exchange packages. Notable uses include:

  Read more →
• 

  Articulatory speech recognition

  Articulatory speech recognition means the recovery of speech (in forms of phonemes, syllables or words) from acoustic signals with the help of articulatory modeling or an extra input of articulatory movement data. Speech recognition (or automatic speech recognition, acoustic speech recognition) means the recovery of speech from acoustics (sound wave) only. Articulatory information is extremely helpful when the acoustic input is in low quality, perhaps because of noise or missing data. Measurable information from the articulatory system (e.g. tongue, jaw movements) can supplement acoustic signals to improve phone recognition accuracy by 2%. However, attempts to estimate articulatory data from acoustic signals alone have not significantly enhanced recognition performance.

  Read more →
• 

  Xara Designer Pro+

  Xara Designer Pro+ is an image editing program incorporating photo editing and vector illustration tools created by British software company Xara. Xara Xtreme LX was an early open source version for Linux. The Windows version was previously sold under the names Xara Studio, Xara X and Xara Xtreme, and traces its origin in the late 1980s to a title called ArtWorks for the Acorn Archimedes line of computers using RISC OS. There is a pro version called Xara Designer Pro (formerly Xara Xtreme Pro). The current commercial version of Xara Photo & Graphic Designer runs only on Windows, although Xara documents can be edited in a web browser on any platform using the Xara Cloud service. Versions up to 4.x can be run on Linux using Wine. == History == ArtWorks, the predecessor of Xara Photo and Graphic Designer, was developed on Acorn Archimedes and Risc PC 32-bit RISC computers running RISC OS by Computer Concepts during the late 1980s. The first version, developed for Microsoft Windows was initially called Xara Studio. It was licensed to Corel Corporation before wide-scale public availability, and from 1995 to 2000 was released as CorelXARA. Corel ceded the licensing rights back to Xara in 2000. The first Xara X version released in 2000 by its original owner. The next version, Xara X¹, was released in 2004. Xara Xtreme was released in 2005. In November 2006, Xara Xtreme PRO (an enhanced version of Xara Xtreme) was released. Xara Xtreme 3.2 and Xtreme Pro 3.2 were released in May 2007. 3.2 Pro included Xara3D, and both versions had more robust typography. In April 2008, Xara Xtreme 4.0 was released. Xara Xtreme and Xara Xtreme Pro 5.1 were released in June 2009. Features included more text-area enhancements, content-aware scaling of bitmap images, improved file import and export, master-page (repeated) objects, an object gallery (replacing the layer gallery), website-creation tools, and multi-stage graduated transparency. In June 2010, Xara Photo & Graphic Designer 6 and Xara Designer Pro 6 were released. Xtreme was renamed Photo & Graphic Designer, and Xtreme Pro was renamed Designer Pro. In May 2011, Xara Photo & Graphic Designer 7 and Xara Designer Pro 7 were released. Features included "magic" photo erase, user interface improvements to docking galleries and snapping alignment, and (in Pro) new webpage and website-design features. In May 2012, Xara Photo & Graphic Designer 2013 and Xara Designer Pro X (v8) were released. Xara Photo & Graphic Designer 9 was released in May 2013. In July of that year, Xara Designer Pro X9 was released. Xara Photo & Graphic Designer 10 was released on 16 July 2014, and Xara Designer Pro X10 on 23 July. Xara Photo & Graphic Designer 11 was released on 29 June 2015, and Xara Designer Pro X11 was released the following month. In 2016, the delivery model was changed to an update service which can be renewed annually. Users are entitled to any updates released while the update service is active. The first update-service updates were in May 2016 for Xara Photo & Graphic Designer, and July 2016 for Xara Designer Pro X. == Features == Xara Photo & Graphic Designer is known for its usability and fast renderer. It provides a fully anti-aliased display, advanced gradient fill, and transparency tools. Among vector editors, Xara Photo & Graphic Designer is considered to be fairly easy to learn, with similarities to CorelDRAW and Inkscape in terms of interface. Alongside the vector illustration tools, Xara Photo & Graphic Designer also includes an integrated photo tool offering manual and automatic photo enhance, cropping, adjustment of brightness levels, red-eye fix, 'magic' erase, photo healing, color and background erase, panoramas and content aware resizing. Designer Pro includes a wider range of tools for other design tasks including the creation of web pages and websites, and text and page layout tools for DTP with the aim of providing a single solution for all graphic and web design tasks.

  Read more →
• 

  2024 National Public Data breach

  In August 2024, three class-action lawsuits were filed against National Public Data along with over 14 complaints filed in federal court, claiming that the company permitted hackers to steal sensitive private information covering millions of individuals. The theft was alleged to have occurred in April 2024. One of the lawsuits specifically claims that in April, a hacker going by the moniker "USDoD" posted a notice on the dark web, offering the data for sale at the price of US$3.5 million. The information stolen is alleged to include 2.9 billion records containing full names, current and past addresses, Social Security numbers, dates of birth, and telephone numbers. The stolen data contains records for people in the US, UK, and Canada. National Public Data confirmed on August 16, 2024, there was a breach originating from someone trying to breach their systems since December 2023, with the breach occurring from April 2024 and over the next few months. The company also confirmed that 2.9 billion records were obtained, though they were still working to determine how many people were affected by the breach, and were working with law enforcement to identify the hacker. == Jerico Pictures == Jerico Pictures, Inc., doing business as National Public Data, was a data broker company that performed employee background checks. Their primary service was collecting information from public data sources, including criminal records, addresses, and employment history, and offering that information for sale. On October 2, 2024, Jerico Pictures filed for Chapter 11 bankruptcy as it currently faces over a dozen lawsuits over the breach, and is potentially liable "for credit monitoring for hundreds of millions of potentially impacted individuals." In December 2024, National Public Data shut down, showing a closure notice on its website.

  Read more →
• 

  Aseprite

  Aseprite ( ace-prite) is a proprietary, source-available image editor designed primarily for pixel art drawing and animation. It runs on Windows, macOS, and Linux, and features different tools for image and animation editing such as layers, frames, tilemap support, command-line interface, Lua scripting, among others. It is developed by Igara Studio S.A. and led by the developers David, Gaspar, and Martín Capello. Aseprite can be downloaded as freeware, (albeit it does not have the ability to save sprites) or purchased on Steam or Itch.io. Aseprite source code and binaries are distributed under EULA, educational, and Steam proprietary licenses. == History == Aseprite, formerly known as Allegro Sprite Editor, had its first release in 2001 as a free software project under the GPLv2 license. This license was kept until August 2016 with version v1.1.8, when the developers switched to a EULA, thus making the software proprietary. On the 1st of September 2016, the main developer, David Capello, wrote a post on the Aseprite Devblog explaining this change. The EULA permits others to download the Aseprite source code, compile it, and use it for personal purposes, but forbids its redistribution to third parties. After the license change, LibreSprite, a free and open source version of it, was created. Both before and after the license change, Aseprite was sold online, on Steam, itch.io, and the project's website. The project's code repository was hosted on Google Code until August 2014, when it was migrated to GitHub, where it remains hosted to date. As of October 2022, its repository has had 68 contributors and around 19 thousand stars. From 2014 to 2021, Aseprite had 66 different releases. Aseprite was used in the development of several notable games such as TowerFall (2013), Celeste (2018), Minit (2018), Wargroove (2019), Loop Hero (2021), Eastward (2021), Unpacking (2021), Haiku the Robot (2022) and Pizza Tower (2023). == Design and features == The main design purpose of Aseprite is to create animated 2D pixel-art sprites. Some of its features include: Layers and frames, with layer grouping and animation tagging Pixel-art specific transformations and tools (pixel-perfect modes, custom brushes, etc.) Animation real-time preview and onion skinning Tilemap and tileset modes Color palette managing, including 65 default palettes Color profiles and modes (RGBA, indexed and grayscale) Non-square pixels Command line interface (CLI) and Lua scripting Aseprite uses its own binary file type to store data, which is typically saved with .ase or .aseprite extensions. Different third-party projects were developed to support parsing of .ase files in programming languages including C#, Python and JavaScript, and in game engines such as Unity and Godot. Images and animations can be exported to different file formats including PNG, GIF, FLC, FLI, JPEG, PCX, TGA, ICO, SVG, and bitmap (BMP).

  Read more →
• 

  Freemake Video Converter

  Freemake Video Converter is a freemium video editing app developed by Ellora Assets Corporation. Designed primarily for entry-level users, the software offers a range of functionalities including video format conversion, DVD ripping, and the creation of photo slideshows and music visualizations. Additionally, Freemake Video Converter is capable of burning video streams that are compatible with various media, such as DVDs and Blu-ray Discs. It also features direct video uploading capabilities to platforms like YouTube., enhancing its utility for content creators. The application's user-friendly interface and broad compatibility make it accessible for individuals with minimal video editing experience. == Features == Freemake Video Converter can perform simple non-linear video editing tasks, such as cutting, rotating, flipping, and combining multiple videos into one file with transition effects. It can also create photo slideshows with background music. Users are then able to upload these videos to YouTube. Freemake Video Converter can read the majority of video, audio, and image formats, and outputs them to AVI, MP4, WMV, Matroska, FLV, SWF, 3GP, DVD, Blu-ray, MPEG and MP3. The program also prepares videos supported by various multimedia devices, including Apple devices (iPod, iPhone, iPad), Xbox, Sony PlayStation, Samsung, Nokia, BlackBerry, and Android mobile devices. The software is able to perform DVD burning and is able to convert videos, photographs, and music into DVD video. The user interface is based on Windows Presentation Foundation technology. Freemake Video Converter supports NVIDIA CUDA technology for H.264 video encoding (starting with version 1.2.0). == Important updates == Freemake Video Converter 2.0 was a major update that integrated two new functions: ripping video from online portals and Blu-ray disc creation and burning. Version 2.1 implemented suggestions from users, including support for subtitles, ISO image creation, and DVD to DVD/Blu-ray conversion. With version 2.3 (earlier 2.2 Beta), support for DXVA has been added to accelerate conversion (up to 50% for HD content). Version 3.0 added HTML5 video creation support and new presets for smartphones. Version 4.0 (introduced in April 2013) added a freemium "Gold Pack" of extra features that can be added if a "donation" is paid. Starting with version 4.0.4, released on 27 August 2013, the program adds a promotional watermark at the end of every video longer than 5 minutes unless Gold Pack is activated. Version 4.1.9, released on 25 November 2015 added support for drag-and-drop functions that were not available in prior versions. Since at least version 4.1.9.44 (1 May 2017), the Freemake Welcome Screen is added at the beginning of the video, and the big Freemake logo is watermarked in the center of the whole video. This decreases the quality of free outputs, and users are forced to pay money to remove the watermark or stop using it. Version 4.1.9.31 (11 August 2016) does not have this restriction. == Licensing issues == FFmpeg has added Freemake Video Converter v1.3 to its Hall of Shame. An issue tracker entry for this product, opened on 16 December 2010, says it is in violation of the GNU General Public License as it is distributing components of the FFmpeg project without including due credit. Ellora Assets Corporation has not responded yet. == Bundled software from sponsors == Since version 4.0, Freemake Video Converter's installer includes a potentially unwanted search toolbar from Conduit as well as SweetPacks malware. Although users can decline the software during installation, the opt-out option is rendered in gray, which could mistakenly give the impression that it's disabled.

  Read more →