Rendezvous hashing

Rendezvous or highest random weight (HRW) hashing is an algorithm that allows clients to achieve distributed agreement on a set of k {\displaystyle k} options out of a possible set of n {\displaystyle n} options. A typical application is when clients need to agree on which sites (or proxies) objects are assigned to. Consistent hashing addresses the special case k = 1 {\displaystyle k=1} using a different method. Rendezvous hashing is both much simpler and more general than consistent hashing (see below). == History == Rendezvous hashing was invented by David Thaler and Chinya Ravishankar at the University of Michigan in 1996. Consistent hashing appeared a year later in the literature. Given its simplicity and generality, rendezvous hashing is now being preferred to consistent hashing in real-world applications. Rendezvous hashing was used very early on in many applications including mobile caching, router design, secure key establishment, and sharding and distributed databases. Other examples of real-world systems that use Rendezvous Hashing include the GitHub load balancer, the Apache Ignite distributed database, the Tahoe-LAFS file store, the CoBlitz large-file distribution service, Apache Druid, IBM's Cloud Object Store, the Arvados Data Management System, Apache Kafka, and the Twitter EventBus pub/sub platform. One of the first applications of rendezvous hashing was to enable multicast clients on the Internet (in contexts such as the MBONE) to identify multicast rendezvous points in a distributed fashion. It was used in 1998 by Microsoft's Cache Array Routing Protocol (CARP) for distributed cache coordination and routing. Some Protocol Independent Multicast routing protocols use rendezvous hashing to pick a rendezvous point. == Problem definition and approach == === Algorithm === Rendezvous hashing solves a general version of the distributed hash table problem: We are given a set of n {\displaystyle n} sites (servers or proxies, say). How can any set of clients, given an object O {\displaystyle O} , agree on a k-subset of sites to assign to O {\displaystyle O} ? The standard version of the problem uses k = 1. Each client is to make its selection independently, but all clients must end up picking the same subset of sites. This is non-trivial if we add a minimal disruption constraint, and require that when a site fails or is removed, only objects mapping to that site need be reassigned to other sites. The basic idea is to give each site S j {\displaystyle S_{j}} a score (a weight) for each object O i {\displaystyle O_{i}} , and assign the object to the highest scoring site. All clients first agree on a hash function h ( ⋅ ) {\displaystyle h(\cdot )} . For object O i {\displaystyle O_{i}} , the site S j {\displaystyle S_{j}} is defined to have weight w i , j = h ( O i , S j ) {\displaystyle w_{i,j}=h(O_{i},S_{j})} . Each client independently computes these weights w i , 1 , w i , 2 … w i , n {\displaystyle w_{i,1},w_{i,2}\dots w_{i,n}} and picks the k sites that yield the k largest hash values. The clients have thereby achieved distributed k {\displaystyle k} -agreement. If a site S {\displaystyle S} is added or removed, only the objects mapping to S {\displaystyle S} are remapped to different sites, satisfying the minimal disruption constraint above. The HRW assignment can be computed independently by any client, since it depends only on the identifiers for the set of sites S 1 , S 2 … S n {\displaystyle S_{1},S_{2}\dots S_{n}} and the object being assigned. HRW easily accommodates different capacities among sites. If site S k {\displaystyle S_{k}} has twice the capacity of the other sites, we simply represent S k {\displaystyle S_{k}} twice in the list, say, as S k , 1 , S k , 2 {\displaystyle S_{k,1},S_{k,2}} . Clearly, twice as many objects will now map to S k {\displaystyle S_{k}} as to the other sites. === Properties === Consider the simple version of the problem, with k = 1, where all clients are to agree on a single site for an object O. Approaching the problem naively, it might appear sufficient to treat the n sites as buckets in a hash table and hash the object name O into this table. Unfortunately, if any of the sites fails or is unreachable, the hash table size changes, forcing all objects to be remapped. This massive disruption makes such direct hashing unworkable. Under rendezvous hashing, however, clients handle site failures by picking the site that yields the next largest weight. Remapping is required only for objects currently mapped to the failed site, and disruption is minimal. Rendezvous hashing has the following properties: Low overhead: The hash function used is efficient, so overhead at the clients is very low. Load balancing: Since the hash function is randomizing, each of the n sites is equally likely to receive the object O. Loads are uniform across the sites. Site capacity: Sites with different capacities can be represented in the site list with multiplicity in proportion to capacity. A site with twice the capacity of the other sites will be represented twice in the list, while every other site is represented once. High hit rate: Since all clients agree on placing an object O into the same site SO, each fetch or placement of O into SO yields the maximum utility in terms of hit rate. The object O will always be found unless it is evicted by some replacement algorithm at SO. Minimal disruption: When a site fails, only the objects mapped to that site need to be remapped. Disruption is at the minimal possible level. Distributed k-agreement: Clients can reach distributed agreement on k sites simply by selecting the top k sites in the ordering. == O(log n) running time via skeleton-based hierarchical rendezvous hashing == The standard version of Rendezvous Hashing described above works quite well for moderate n, but when n {\displaystyle n} is extremely large, the hierarchical use of Rendezvous Hashing achieves O ( log ⁡ n ) {\displaystyle O(\log n)} running time. This approach creates a virtual hierarchical structure (called a "skeleton"), and achieves O ( log ⁡ n ) {\displaystyle O(\log n)} running time by applying HRW at each level while descending the hierarchy. The idea is to first choose some constant m {\displaystyle m} and organize the n {\displaystyle n} sites into c = ⌈ n / m ⌉ {\displaystyle c=\lceil n/m\rceil } clusters C 1 = { S 1 , S 2 … S m } , C 2 = { S m + 1 , S m + 2 … S 2 m } … {\displaystyle C_{1}=\left\{S_{1},S_{2}\dots S_{m}\right\},C_{2}=\left\{S_{m+1},S_{m+2}\dots S_{2m}\right\}\dots } Next, build a virtual hierarchy by choosing a constant f {\displaystyle f} and imagining these c {\displaystyle c} clusters placed at the leaves of a tree T {\displaystyle T} of virtual nodes, each with fanout f {\displaystyle f} . In the accompanying diagram, the cluster size is m = 4 {\displaystyle m=4} , and the skeleton fanout is f = 3 {\displaystyle f=3} . Assuming 108 sites (real nodes) for convenience, we get a three-tier virtual hierarchy. Since f = 3 {\displaystyle f=3} , each virtual node has a natural numbering in octal. Thus, the 27 virtual nodes at the lowest tier would be numbered 000 , 001 , 002 , . . . , 221 , 222 {\displaystyle 000,001,002,...,221,222} in octal (we can, of course, vary the fanout at each level - in that case, each node will be identified with the corresponding mixed-radix number). The easiest way to understand the virtual hierarchy is by starting at the top, and descending the virtual hierarchy. We successively apply Rendezvous Hashing to the set of virtual nodes at each level of the hierarchy, and descend the branch defined by the winning virtual node. We can in fact start at any level in the virtual hierarchy. Starting lower in the hierarchy requires more hashes, but may improve load distribution in the case of failures. For example, instead of applying HRW to all 108 real nodes in the diagram, we can first apply HRW to the 27 lowest-tier virtual nodes, selecting one. We then apply HRW to the four real nodes in its cluster, and choose the winning site. We only need 27 + 4 = 31 {\displaystyle 27+4=31} hashes, rather than 108. If we apply this method starting one level higher in the hierarchy, we would need 9 + 3 + 4 = 16 {\displaystyle 9+3+4=16} hashes to get to the winning site. The figure shows how, if we proceed starting from the root of the skeleton, we may successively choose the virtual nodes ( 2 ) 3 {\displaystyle (2)_{3}} , ( 20 ) 3 {\displaystyle (20)_{3}} , and ( 200 ) 3 {\displaystyle (200)_{3}} , and finally end up with site 74. The virtual hierarchy need not be stored, but can be created on demand, since the virtual nodes names are simply prefixes of base- f {\displaystyle f} (or mixed-radix) representations. We can easily create appropriately sorted strings from the digits, as required. In the example, we would be working with the strings 0 , 1 , 2 {\displaystyle 0,1,2} (at tier 1), 20 , 21 , 22 {\displaystyle 20,21,22} (at tier 2), and 200 , 201 , 202

Artificial general intelligence

Artificial general intelligence (AGI) is a hypothetical type of artificial intelligence that matches or surpasses human capabilities across virtually all cognitive tasks. Beyond AGI, artificial superintelligence (ASI) would outperform the best human abilities across every domain by a wide margin. Unlike artificial narrow intelligence (ANI), whose competence is confined to well‑defined tasks, an AGI system can generalise knowledge, transfer skills between domains, and solve novel problems without task‑specific reprogramming. Creating AGI is a stated goal of technology companies such as OpenAI, Google, xAI, and Meta. A 2020 survey identified 72 active AGI research and development projects across 37 countries. AGI is a common topic in science fiction and futures studies. Contention exists over whether AGI represents an existential risk. Some AI experts and industry figures have stated that mitigating the risk of human extinction posed by AGI should be a global priority. Others find the development of AGI to be in too remote a stage to present such a risk. == Terminology == AGI is also known as strong AI, full AI, human-level AI, human-level intelligent AI, or general intelligent action. The term "artificial general intelligence" was used in 1997 by Mark Gubrud in a discussion of the implications of fully automated military production and operations. A mathematical formalism of AGI named AIXI was proposed in 2000 by Marcus Hutter, who defines intelligence as "an agent’s ability to achieve goals or succeed in a wide range of environments". This type of AGI has also been called "universal artificial intelligence". The term AGI was re-introduced and popularized by Shane Legg and Ben Goertzel around 2002. Some academic sources reserve the term "strong AI" for computer programs that will experience sentience or consciousness. In contrast, weak AI (or narrow AI) can solve a specific problem but lacks general cognitive abilities. Some academic sources use "weak AI" to refer more broadly to any programs that neither experience consciousness nor have a mind in the same sense as humans. Related concepts include artificial superintelligence and transformative AI. An artificial superintelligence (ASI) is a hypothetical type of AGI that is much more generally intelligent than humans, while the notion of transformative AI relates to AI having a large impact on society, for example, similar to the agricultural or industrial revolution. A framework for classifying AGI was proposed in 2023 by Google DeepMind researchers. They define five performance levels of AGI: emerging, competent, expert, virtuoso, and superhuman. For example, a competent AGI is defined as an AI that outperforms 50% of skilled adults in a wide range of non-physical tasks, and a superhuman AGI (i.e., an artificial superintelligence) is similarly defined but with a threshold of 100%. They consider large language models like ChatGPT or LLaMA 2 to be instances of emerging AGI (comparable to unskilled humans). Regarding the autonomy of AGI and associated risks, they define five levels: tool (fully in human control), consultant, collaborator, expert, and agent (fully autonomous). == Characteristics == There is no single agreed-upon definition of intelligence as applied to computers. Computer scientist John McCarthy wrote in 2007: "We cannot yet characterize in general what kinds of computational procedures we want to call intelligent." === Intelligence traits === Researchers generally hold that a system is required to do all of the following to be regarded as an AGI: reason, use strategy, solve puzzles, and make judgments under uncertainty, represent knowledge, including common sense knowledge, plan, learn, communicate in natural language, if necessary, integrate these skills in completion of any given goal. Many interdisciplinary approaches (e.g. cognitive science, computational intelligence, and decision making) consider additional traits such as imagination (the ability to form novel mental images and concepts) and autonomy. Computer-based systems exhibiting these capabilities are now widespread, with modern large language models demonstrating computational creativity, automated reasoning, and decision support simultaneously across domains. === Physical traits === Other capabilities are considered desirable in intelligent systems, as they may affect intelligence or aid in its expression. These include: the ability to sense (e.g. see, hear, etc.), and the ability to act (e.g. move and manipulate objects, change location to explore, etc.) This includes the ability to detect and respond to hazard. === Tests for human-level AGI === Several tests meant to confirm human-level AGI have been considered. ==== Turing test ==== The Turing test was proposed by Alan Turing in his 1950 paper "Computing Machinery and Intelligence". This test involves a human judge engaging in natural language conversations with both a human and a machine designed to generate human-like responses. The machine passes the test if it can convince the judge that it is human a significant fraction of the time. Turing proposed this as a practical measure of machine intelligence, focusing on the ability to produce human-like responses rather than on the internal workings of the machine. The idea of the test is that the machine has to try and pretend to be a man, by answering questions put to it, and it will only pass if the pretence is reasonably convincing. A considerable portion of a jury, who should not be experts about machines, must be taken in by the pretence. In 2014, a chatbot named Eugene Goostman, designed to imitate a 13-year-old Ukrainian boy, reportedly passed a Turing Test event by convincing 33% of judges that it was human. However, this claim was met with significant skepticism from the AI research community, who questioned the test's implementation and its relevance to AGI. A 2025 pre‑registered, three‑party Turing‑test study by Cameron R. Jones and Benjamin K. Bergen showed that GPT-4.5 was judged to be the human in 73% of five‑minute text conversations—surpassing the 67% humanness rate of real confederates and meeting the researchers' criterion for having passed the test. ==== Ikea test ==== The "Ikea test", also known as the Flat Pack Furniture Test, involves an AI controlling a robot which attempts to assemble an Ikea flat-pack furniture product after having been shown the parts and instructions. As early as 2013, MIT's IkeaBot demonstrated fully autonomous multi-robot assembly of an IKEA Lack table in ten minutes, with no human intervention and no pre-programmed assembly instructions. The robots inferred the assembly sequence from the geometry of the parts alone. ==== Coffee test ==== Steve Wozniak proposed a test where a machine is required to enter an average American home and figure out how to make coffee. It must find the coffee machine, find the coffee, add water, find a mug, and brew the coffee by pushing the proper buttons. This test has been substantially approached across multiple systems. In January 2024, Figure AI's Figure 01 humanoid learned to operate a Keurig coffee machine autonomously after watching video demonstrations, using end-to-end neural networks to translate visual input into motor actions. In 2025, researchers at the University of Edinburgh published the ELLMER framework in Nature Machine Intelligence, demonstrating a robotic arm that interprets verbal instructions, analyses its surroundings, and autonomously makes coffee in dynamic kitchen environments — adapting to unforeseen obstacles in real time rather than following pre-programmed sequences. ==== Suleyman's test ==== Mustafa Suleyman's test proposes giving an AI model US$100,000 and asking it to obtain US$1 million. ==== Use of video-games ==== Adams, et al. propose that the ability to learn and succeed in a wide range of video games can be used to test AI intelligence. This range would include games unknown to the AGI developers before the test is administered. === AI-complete problems === A problem is informally called "AI-complete" or "AI-hard" if it is believed that AGI would be needed to solve it, because the solution is beyond the capabilities of a purpose-specific algorithm. == History == === Classical AI === Modern AI research began in the mid-1950s. The first generation of AI researchers were convinced that artificial general intelligence was possible and that it would exist in just a few decades. AI pioneer Herbert A. Simon wrote in 1965: "machines will be capable, within twenty years, of doing any work a man can do". Their predictions were the inspiration for Stanley Kubrick and Arthur C. Clarke's fictional character HAL 9000, who embodied what AI researchers believed they could create by the year 2001. AI pioneer Marvin Minsky was a consultant on the project of making HAL 9000 as realistic as possible according to the consensus predictions of the time. He said in 1967, "Within a generation... the problem of

Macroelectronics

Macroelectronics are flexible electronics that cover a large area. The most visible example of macroelectronics is flat-panel displays. Other emerging applications include rollable display, printable thin film solar cell and electronic skin. Flat-panel displays fabricated on glass substrates are fragile so fabricating directly on flexible substrates, such as polymers is being explored. Displays made on thin polymer substrates can be more rugged than glass. In September 2005, Philips Polymer Vision revealed the world's first prototype of a rollable electronic reader, which can unfold to a 5-inch display and roll back into a pocket-size (100×60×20 mm) device. Thin-film devices on flexible polymer substrates can lend themselves to low-cost fabrication processes (i.e., roll-to-roll printing), resulting in lightweight, rugged and flexible macroelectronic products.

Open Sound Control

Open Sound Control (OSC) is a protocol for networking sound synthesizers, computers, and other multimedia devices for purposes such as musical performance or show control. OSC's advantages include interoperability, accuracy, flexibility and enhanced organization and documentation. Its disadvantages include higher bandwidth requirements, increased load on embedded processors, and lack of standardized messages/interoperability. The first specification was released in March 2002. == Motivation == OSC is a content format developed at CNMAT by Adrian Freed and Matt Wright comparable to XML, WDDX, or JSON. It was originally intended for sharing music performance data (gestures, parameters and note sequences) between musical instruments (especially electronic musical instruments such as synthesizers), computers, and other multimedia devices. OSC is sometimes used as an alternative to the 1983 MIDI standard, when higher resolution and a richer parameter space is desired. OSC messages are transported across the internet and within local subnets using UDP/IP and Ethernet. OSC messages between gestural controllers are usually transmitted over serial endpoints of USB wrapped in the SLIP protocol. == Features == OSC's main features, compared to MIDI, include: Open-ended, dynamic, URI-style symbolic naming scheme Symbolic and high-resolution numeric data Pattern matching language to specify multiple recipients of a single message High resolution time tags "Bundles" of messages whose effects must occur simultaneously == Applications == There are dozens of OSC applications, including real-time sound and media processing environments, web interactivity tools, software synthesizers, programming languages and hardware devices. OSC has achieved wide use in fields including musical expression, robotics, video performance interfaces, distributed music systems and inter-process communication. The TUIO community standard for tangible interfaces such as multitouch is built on top of OSC. Similarly the GDIF system for representing gestures integrates OSC. OSC is used extensively in experimental musical controllers, and has been built into several open source and commercial products. The Open Sound World (OSW) music programming language is designed around OSC messaging. OSC is the heart of the DSSI plugin API, an evolution of the LADSPA API, in order to make the eventual GUI interact with the core of the plugin via messaging the plugin host. LADSPA and DSSI are APIs dedicated to audio effects and synthesizers. In 2007, a standardized namespace within OSC called SYN, for communication between controllers, synthesizers and hosts, was proposed. == Design == OSC messages consist of an address pattern (such as /oscillator/4/frequency), a type tag string (such as ,fi for a float32 argument followed by an int32 argument), and the arguments themselves (which may include a time tag). Address patterns form a hierarchical name space, reminiscent of a Unix filesystem path, or a URL, and refer to "Methods" inside the server, which are invoked with the attached arguments. Type tag strings are a compact string representation of the argument types. Arguments are represented in binary form with four-byte alignment. The core types supported are 32-bit two's complement signed integers 32-bit IEEE floating point numbers Null-terminated arrays of eight-bit encoded data (C-style strings) arbitrary sized blob (e.g. audio data, or a video frame) An example message is included in the spec (with null padding bytes represented by ␀): /oscillator/4/frequency␀,f␀␀, Followed by the 4-byte float32 representation of 440.0: 0x43dc0000. Messages may be combined into bundles, which themselves may be combined into bundles, etc. Each bundle contains a timestamp, which determines whether the server should respond immediately or at some point in the future. Applications commonly employ extensions to this core set. More recently some of these extensions such as a compact Boolean type were integrated into the required core types of OSC 1.1. The advantages of OSC over MIDI are primarily internet connectivity; data type resolution; and the comparative ease of specifying a symbolic path, as opposed to specifying all connections as seven-bit numbers with seven-bit or fourteen-bit data types. This human-readability has the disadvantage of being inefficient to transmit and more difficult to parse by embedded firmware, however. The spec does not define any particular OSC Methods or OSC Containers. All messages are implementation-defined and vary from server to server.

Algorithmic curation

Algorithm curation is the selection of online media by technologies such as recommender systems and personalized search. Curation entails the selective sharing of online content and recommendations based on inferred interests. Curation algorithms implement different filter approaches, such as collaborative filtering and content-based filtering. Examples include search engine and social media products such as the Twitter feed, Facebook's News Feed, and Google Personalized Search. == History == === Early algorithmic curation === Online platforms use newsfeed algorithms to determine what content to present to each user. The volume of content published on social media platforms created a need for automated filtering, as manual review of all available content by users is not feasible. These systems function as a form of gatekeeper, shaping which new material users are exposed to and influencing knowledge, attention, and political exposure. ==== Information overload ==== Early ranking algorithms addressed information overload by surfacing the most recent or most popular posts. Later systems shifted toward ranking content based on predicted engagement, aiming to increase the time users spend on a platform. Research has found that these engagement-oriented systems can increase the spread of misinformation and contribute to political polarization as a side effect of optimising for user interaction. ==== How algorithm changes users' feeds over time ==== Algorithmic curation has been found to increase source diversity in some respects while simultaneously reducing the number of external links presented to users, which limits exposure to off-platform content. Research using agent-based modelling has examined how user behaviour, information quality, and algorithmic design interact with one another over time. === Emergence of AI === Platforms increasingly shifted from rule-based ranking systems toward machine-learning and AI-driven approaches, which allow feeds to be personalised at a larger scale and with greater responsiveness to user behaviour. For example, X (formerly Twitter) moved away from a chronological feed toward an AI-powered ranking system that personalises content for each user. These systems are capable of making ranking decisions across volumes of content and user interactions that would not be practical to handle manually. == Approach == === Filter types === ==== Collaborative filtering ==== Collaborative filtering (CF) methods create recommendations based on a person's usage patterns. CF predicts a person's preference for an item by matching their interests with those of users who have similar interests. This process allows for the sharing of ratings between users with similar profiles. CF is based on patterns of human behaviour rather than machine analysis of content itself. Users of CF systems rate items they have interacted with, and these ratings form a profile of interests. The CF system then matches that user with others who have similar profiles, and uses their ratings to generate recommendations. Collaborative filtering can be applied across various content types including text, images, music, and financial products, and can account for complex attributes such as taste and quality that are difficult to represent explicitly. ==== Content-based filtering ==== Content-based filtering (CBF) builds a user profile to represent the types of items a user has engaged with, based on keywords and attributes used to describe those items. Recommendations are generated by presenting items similar to those the user has previously engaged with or is currently viewing. The CBF method creates a profile for each item based on discrete attributes and features, and then constructs a content-based user profile using a weighted vector of those features derived from items the user has rated, purchased, or interacted with. The weights represent the relative importance of each feature, and can be computed using techniques such as Bayesian classifiers, cluster analysis, decision trees, and artificial neural networks, with the goal of estimating the probability that a user will engage with a suggested item. One application of content-based filtering is Pandora Radio, where users provide an artist, genre, or composer to generate a station, and the system surfaces music with similar attributes. == Technology == === Recommender system === Recommender systems rank and suggest content to users based on a combination of implicit and explicit user input. Implicit signals include time spent viewing or engaging with a specific item. Explicit signals include actions such as liking posts, saving store pages, reading news articles, or sharing content. === Personalized search === Personalized search aims to retrieve results most relevant to the user by incorporating contextual factors beyond the explicit query, such as past queries, browsing history, and inferred interests. Social media platforms such as X (formerly Twitter) and Bluesky generate recommendations based on similar users and the content those users interact with. Personalized search may also allow users to explicitly filter results by blocking content containing certain phrases or hashtags. For first-time users without prior history, personalized search may draw on content-based filtering to establish an initial context. Similar processes are used by search engines and retail platforms to tailor results and product recommendations to individual users. == AI contribution == Artificial intelligence contributes to algorithmic curation through machine-learning models capable of processing large volumes of data. Techniques such as deep learning and reinforcement learning allow curation algorithms to model user preferences with greater granularity alongside established filtering approaches. This enables platforms to adjust content rankings rapidly in response to user behaviour. In social media and streaming contexts, AI-driven systems arrange feeds according to predicted relevance, with the outputs shaped by patterns present in the training data. == Social media and potential impact == === Echo chambers === Social media algorithms, such as those used by X (formerly Twitter), recommend content that the system predicts a user will engage with positively. Content from accounts with differing perspectives is less likely to be surfaced, which may reduce source and topic diversity and contribute to the formation of echo chambers. For example, Facebook's news feed is designed to surface content aligned with users' prior engagement, which may reinforce existing views. This dynamic may contribute to filter bubbles, in which users are seldom exposed to content outside their existing interests. Users may further narrow their feeds by actively blocking certain content or accounts. === Over-representation === A pattern observed across social media platforms is the concentration of algorithmic visibility among a small subset of users. Content from the most active users, those with the largest followings, or those generating the most engagement tends to be surfaced more frequently, meaning a small number of accounts can account for a disproportionate share of what appears in other users' feeds.

Outline of deep learning

The following outline is provided as an overview of, and topical guide to, deep learning: Deep learning is a subfield of machine learning and artificial intelligence based on artificial neural networks with multiple processing layers. It emphasizes representation learning and is widely used in areas such as computer vision, natural language processing, speech recognition, recommender systems, robotics, and generative artificial intelligence. == Ways to categorize deep learning == A field of study A branch of artificial intelligence A subfield of machine learning A subfield of computer science A form of representation learning A class of methods based on artificial neural networks An approach used in computational statistics == History == === Precursors === Cybernetics Perceptron Connectionism Neocognitron Backpropagation === Milestones === LeNet Long short-term memory Deep belief network AlexNet Sequence to sequence learning Generative adversarial network Residual neural network Transformer BERT Generative pre-trained transformer Diffusion model === Related histories === History of artificial intelligence History of machine learning Timeline of machine learning == Core concepts == == Learning settings == Supervised learning Unsupervised learning Self-supervised learning Semi-supervised learning Reinforcement learning Transfer learning Multitask learning Multimodal learning Online machine learning Continual learning == Common tasks == Image classification Object detection Image segmentation Automatic speech recognition Neural machine translation Question answering Automatic summarization Text-to-image model Protein structure prediction == Architectures == === Feedforward and convolutional architectures === Feedforward neural network Multilayer perceptron Convolutional neural network Radial basis function network Residual neural network U-Net === Recurrent and sequence architectures === Recurrent neural network Long short-term memory Gated recurrent unit Sequence to sequence learning Recursive neural network === Representation-learning architectures === Autoencoder Denoising autoencoder Sparse autoencoder Variational autoencoder Restricted Boltzmann machine Deep belief network === Attention and transformer architectures === Attention (machine learning) Transformer BERT Generative pre-trained transformer Vision transformer === Generative and probabilistic architectures === Autoregressive model Diffusion model Energy-based model Generative adversarial network Mixture of experts === Graph and memory architectures === Graph neural network Graph convolutional network Siamese network Neural Turing machine Memory network Echo state network Capsule neural network == Neural network components and techniques == Artificial neuron Activation function Rectified linear unit Sigmoid function Softmax function Embedding Convolution Pooling layer Attention Batch normalization Layer normalization Residual connections == Training and optimization == Backpropagation Gradient descent Stochastic gradient descent Adam optimization Learning rate Loss function Cross-entropy Mean squared error Regularization Dropout Early stopping Batch normalization Data augmentation Transfer learning Knowledge distillation Ensemble learning Curriculum learning == Datasets and benchmarks == CIFAR-10 ImageNet MNIST database Common Objects in Context (COCO) General Language Understanding Evaluation (GLUE) benchmark LibriSpeech SQuAD == Applications == === Computer vision === Computer vision Facial recognition system Image classification Image segmentation Medical imaging Object detection Optical character recognition === Natural language processing === Automatic summarization Chatbot Information retrieval Large language model Natural language processing Neural machine translation Question answering Sentiment analysis === Speech and audio === Automatic speech recognition Music information retrieval Speaker recognition Speech synthesis === Science and medicine === Bioinformatics Computational biology Drug discovery Medical diagnosis Protein structure prediction === Robotics and control === Autonomous car Computer game bot Control theory Robotics === Recommendation, search, and forecasting === Anomaly detection Forecasting Fraud detection Recommender system Search engine === Generative artificial intelligence === Deepfake Generative artificial intelligence Large language model Speech synthesis Text-to-image model === Computer graphics and video games === Deep Learning Anti-Aliasing (DLAA) Deep Learning Super Sampling (DLSS) == Hardware == AMD Instinct AMD XDNA Application-specific integrated circuit Deep learning processor, Neural processing unit (NPU), or Neural Engine Field-programmable gate array General-purpose computing on graphics processing units (GPGPU) Graphics processing unit NVIDIA Deep Learning Accelerator (NVDLA) Tensor processing unit Vision processing unit Wafer-scale integration === Supporting software platforms === CUDA Metal ROCm == Software == === Open-source frameworks and libraries === === Neural network software === EDLUT Emergent Encog JOONE Neuroph NeuroSolutions OpenNN Peltarion Synapse SNNS === Platforms, tools, and deployment === Amazon SageMaker Google Colab Hugging Face Kaggle Kubeflow MLflow ONNX OpenVINO TensorFlow Hub == Algorithms for deep learning and neural networks == Backpropagation Conjugate gradient method Generalized Hebbian algorithm Gradient descent Levenberg–Marquardt algorithm Perceptron Quasi-Newton method Wake-sleep algorithm == Methods and related topics == === Representation and metric learning === Contrastive learning Embedding Feature learning Manifold learning Metric learning === Generative modeling === Autoregressive model Diffusion model Generative adversarial network Generative model Variational inference === Efficient and scalable deep learning === Knowledge distillation Low-rank approximation Mixture of experts Quantization Sparsity === Reliability, safety, and interpretability === Adversarial machine learning AI alignment Algorithmic bias Catastrophic forgetting Differential privacy Explainable artificial intelligence Federated learning Hallucination (artificial intelligence) == Conferences and workshops == Annual Meeting of the Association for Computational Linguistics Conference on Computer Vision and Pattern Recognition Conference on Neural Information Processing Systems International Conference on Computer Vision International Conference on Learning Representations International Conference on Machine Learning == Organizations == === Research laboratories and institutions === Allen Institute for AI Alberta Machine Intelligence Institute European Laboratory for Learning and Intelligent Systems Google DeepMind Meta AI Mila Microsoft Research Vector Institute === Companies === Anthropic Cerebras Cohere DeepSeek Mistral AI OpenAI Stability AI xAI == Publications == === Books === Deep Learning – Ian Goodfellow and Yoshua Bengio Neural Networks and Deep Learning – Michael Nielsen Perceptrons – Marvin Minsky and Seymour Papert === Journals === IEEE Transactions on Neural Networks and Learning Systems Neural Networks Neural Computation == Influential persons ==

International World Wide Web Conference Committee

The International World Wide Web Conference Committee (abbreviated as IW3C2 also written as IW3C2) is a professional non-profit organization registered in Switzerland (Article 60ff of the Swiss Civil Code) that promotes World Wide Web research and development. The IW3C2 organizes and hosts the annual World Wide Web Conference in conjunction with the W3C. The IW3C2 was founded by Joseph Hardin and Robert Cailliau at a meeting held in Boston, United States, on 14 August 1994 to prepare for the upcoming Second International World Wide Web Conference in Chicago. The IW3C2 formally became an incorporated entity in May 1996 at the fifth conference in Paris, France. The organization is governed by laws of the Swiss Confederation and the By-laws. == Abbreviation == The abbreviation for the International World Wide Web Conference Committee as IW3C2 is as follow: I- The I is represents the leading I in International. W3- The W3 represents the three 3 leading W's in World Wide Web. C2- The C2 represents the three 2 leading C's in Conference Committee. == Mission == The mission of the IW3C2 is: To coordinate the organization and planning of the international WWW conference series and ensure that it remains the foremost conference addressing World Wide Web research and development; To promote a collaborative spirit among conference attendees that is essential to the success of the series; To ensure the global geographical diversity of conference sites and provide support to local organizers at those sites; To make sure that all content arising from these conferences and forums is permanently and openly available on the widest possible scale; To preserve the history of the conference series; To encourage the global development of the World Wide Web through collaboration with WWW standards organizations; To provide a permanent, broad-based international body to achieve these purposes. == Conferences == The conferences are organized by the IW3C2 in collaboration with local organizing committees and technical program committees. The series provides an open forum in which all opinions can be presented, subject to a strict process of peer review. The proceedings of the conference are published in the ACM Digital Library. === Endorsed conferences === The IW3C2 has endorsed regional conferences devoted to a special topic of the Web by working with endorsed conferences on cross-promotion, publicity and programs. == Membership == Members of the IW3C2 are ordinary members, ex officio members, non-voting members, and officers. === Ordinary members === Ordinary members are elected for a period of 3 years during a general meeting. Members are nominated due to their recognition in the WWW community and represent themselves. Members can be re-elected only after at least one year of absence. The following are the founding members at the time when IW3C2 was officially incorporated in May 1996: Jean-François Abramatic Tim Berners-Lee Robert Cailliau Dale Dougherty Ira Goldstein Joseph Hardin Tim Krauskopf Detlef Krömker Corinne Moore R. P. Channing Rodgers Albert Vezza Stuart Weibel Yuri Rubinsky (died prior to incorporation) The following are the current (April 2016) ordinary members: Robin Chen Chin-Wan Chung Allan Ellis Wendy Hall - IW3C2 Chair Ivan Herman Arun Iyengar - IW3C2 Vice Chair Irwin King Yoelle Maarek Luc Mariaux - IW3C2 Treasurer Daniel Schwabe - IW3C2 Vice-Chair === Ex officio members === Ex officio members are selected from the immediate past conference general co-chairs and from future conference co-chairs. Their term expires one year after the conference they organized. Ex officio members can be elected as ordinary members. The following are current (April 2016) ex officio members and the conference with which they are affiliated: Jacqueline Bourdeau - WWW2016 James Hendler - WWW2016 Rick Barrett - WWW2017 Rick Cummings - WWW2017 Laurent Flory - WWW2018 Fabien Gandon - WWW2018 === Officers === The IW3C2 officers consist of a chairperson, a vice-chair (chairperson-elect), a secretary, a treasurer, and other appointees. Officers are elected during a general meeting (usually at the annual WWW conference) and serve for one year. They can be re-elected an indefinite number of times. == The Seoul Test of Time Award == This annual award, presented at the WWW conference, is made possible by a generous contribution from the organizers of WWW2014 (Seoul Korea). Recipients are determined by the IW3C2 and honor the author, or authors, of a paper presented at a previous WWW conference that has "stood the test of time." The first award, announced at WWW2015 (Florence Italy), recognized Sergey Brin and Larry Page, the founders of Google. The recipients of the WWW2016 award are LinkIn scientist Dr. Badrul Sarwar and University of Minnesota professors George Karypis, Joseph Konstan, and John Riedl (posthumous) for their work in item-item collaborative filtering.