AI Chatbot Creator

AI Chatbot Creator — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

• 

  Situated approach (artificial intelligence)

  In artificial intelligence research, the situated approach builds agents that are designed to behave effectively successfully in their environment. This requires designing AI "from the bottom-up" by focussing on the basic perceptual and motor skills required to survive. The situated approach gives a much lower priority to abstract reasoning or problem-solving skills. The approach was originally proposed as an alternative to traditional approaches (that is, approaches popular before 1985 or so). After several decades, classical AI technologies started to face intractable issues (e.g. combinatorial explosion) when confronted with real-world modeling problems. All approaches to address these issues focus on modeling intelligences situated in an environment. They have become known as the situated approach to AI. == Emergence of a concept == === From traditional AI to Nouvelle AI === During the late 1980s, the approach now known as Nouvelle AI (Nouvelle means new in French) was pioneered at the MIT Artificial Intelligence Laboratory by Rodney Brooks. As opposed to classical or traditional artificial intelligence, Nouvelle AI purposely avoided the traditional goal of modeling human-level performance, but rather tries to create systems with intelligence at the level of insects, closer to real-world robots. But eventually, at least at MIT new AI did lead to an attempt for humanoid AI in the Cog Project. === From Nouvelle AI to behavior-based and situated AI === The conceptual shift introduced by nouvelle AI flourished in the robotics area, given way to behavior-based robotics (BBR), a methodology for developing AI based on a modular decomposition of intelligence. It was made famous by Rodney Brooks: his subsumption architecture was one of the earliest attempts to describe a mechanism for developing BBAI. It is extremely popular in robotics and to a lesser extent to implement intelligent virtual agents because it allows the successful creation of real-time dynamic systems that can run in complex environments. For example, it underlies the intelligence of the Sony Aibo and many RoboCup robot teams. Realizing that in fact all these approaches were aiming at building not an abstract intelligence, but rather an intelligence situated in a given environment, they have come to be known as the situated approach. In fact, this approach stems out from early insights of Alan Turing, describing the need to build machines equipped with sense organs to learn directly from the real-world instead of focusing on abstract activities, such as playing chess. == Definitions == Classically, a software entity is defined as a simulated element, able to act on itself and on its environment, and which has an internal representation of itself and of the outside world. An entity can communicate with other entities, and its behavior is the consequence of its perceptions, its representations, and its interactions with the other entities. === AI loop === Simulating entities in a virtual environment requires simulating the entire process that goes from a perception of the environment, or more generally from a stimulus, to an action on the environment. This process is called the AI loop and technology used to simulate it can be subdivided in two categories. Sensorimotor or low-level AI deals with either the perception problem (what is perceived?) or the animation problem (how are actions executed?). Decisional or high-level AI deals with the action selection problem (what is the most appropriate action in response to a given perception, i.e. what is the most appropriate behavior?). === Traditional or symbolic AI === There are two main approaches in decisional AI. The vast majority of the technologies available on the market, such as planning algorithms, finite-state machines (FSA), or expert systems, are based on the traditional or symbolic AI approach. Its main characteristics are: It is top-down: it subdivides, in a recursive manner, a given problem into a series of sub-problems that are supposedly easier to solve. It is knowledge-based: it relies on a symbolic description of the world, such as a set of rules. However, the limits of traditional AI, which goal is to build systems that mimic human intelligence, are well-known: inevitably, a combinatorial explosion of the number of rules occurs due to the complexity of the environment. In fact, it is impossible to predict all the situations that will be encountered by an autonomous entity. === Situated or behavioral AI === In order to address these issues, another approach to decisional AI, also known as situated or behavioral AI, has been proposed. It does not attempt to model systems that produce deductive reasoning processes, but rather systems that behave realistically in their environment. The main characteristics of this approach are the following: It is bottom-up: it relies on elementary behaviors, which can be combined to implement more complex behaviors. It is behavior-based: it does not rely on a symbolic description of the environment, but rather on a model of the interactions of the entities with their environment. The goal of situated AI is to model entities that are autonomous in their environment. This is achieved thanks to both the intrinsic robustness of the control architecture, and its adaptation capabilities to unforeseen situations. === Situated agents === In artificial intelligence and cognitive science, the term situated refers to an agent which is embedded in an environment. The term situated is commonly used to refer to robots, but some researchers argue that software agents can also be situated if: they exist in a dynamic (rapidly changing) environment, which they can manipulate or change through their actions, and which they can sense or perceive. Examples might include web-based agents, which can alter data or trigger processes (such as purchases) over the Internet, or virtual-reality bots which inhabit and change virtual worlds, such as Second Life. Being situated is generally considered to be part of being embodied, but it is useful to consider each perspective individually. The situated perspective emphasizes that intelligent behavior derives from the environment and the agent's interactions with it. The nature of these interactions are defined by an agent's embodiment. == Implementation principles == === Modular decomposition === The most important attribute of a system driven by situated AI is that the intelligence is controlled by a set of independent semi-autonomous modules. In the original systems, each module was actually a separate device or was at least conceived of as running on its own processing thread. Generally, though, the modules are just abstractions. In this respect, situated AI may be seen as a software engineering approach to AI, perhaps akin to object oriented design. Situated AI is often associated with reactive planning, but the two are not synonymous. Brooks advocated an extreme version of cognitive minimalism which required initially that the behavior modules were finite-state machines and thus contained no conventional memory or learning. This is associated with reactive AI because reactive AI requires reacting to the current state of the world, not to an agent's memory or preconception of that world. However, learning is obviously key to realistic strong AI, so this constraint has been relaxed, though not entirely abandoned. === Action selection mechanism === The situated AI community has presented several solutions to modeling decision-making processes, also known as action selection mechanisms. The first attempt to solve this problem goes back to subsumption architectures, which were in fact more an implementation technique than an algorithm. However, this attempt paved the way to several others, in particular the free-flow hierarchies and activation networks. A comparison of the structure and performances of these two mechanisms demonstrated the advantage of using free-flow hierarchies in solving the action selection problem. However, motor schemas and process description languages are two other approaches that have been used with success for autonomous robots. == Notes and references == Arsenio, Artur M. (2004) Towards an embodied and situated AI, In: Proceedings of the International FLAIRS conference, 2004. (online) The Artificial Life Route To Artificial Intelligence: Building Embodied, Situated Agents, Luc Steels and Rodney Brooks Eds., Lawrence Erlbaum Publishing, 1995. (ISBN 978-0805815184) Rodney A. Brooks Cambrian Intelligence (MIT Press, 1999) ISBN 0-262-52263-2; collection of early papers including "Intelligence without representation" and "Intelligence without reason", from 1986 & 1991 respectively. Ronald C. Arkin Behavior-Based Robotics (MIT Press, 1998) ISBN 0-262-01165-4 Hendriks-Jansen, Horst (1996) Catching Ourselves in the Act: Situated Activity, Interactive Emergence, Evolution, and Human Thought. Cambridge, Mass.: MIT Press.

  Read more →
• 

  Magic state distillation

  Magic state distillation is a method for creating more accurate quantum states from multiple noisy ones, which is important for building fault tolerant quantum computers. It has also been linked to quantum contextuality, a concept thought to contribute to quantum computers' power. The technique was first proposed by Emanuel Knill in 2004, and further analyzed by Sergey Bravyi and Alexei Kitaev the same year. Thanks to the Gottesman–Knill theorem, it is known that some quantum operations (operations in the Clifford group) can be perfectly simulated in polynomial time on a classical computer. In order to achieve universal quantum computation, a quantum computer must be able to perform operations outside this set. Magic state distillation achieves this, in principle, by concentrating the usefulness of imperfect resources, represented by mixed states, into states that are conducive for performing operations that are difficult to simulate classically. A variety of qubit magic state distillation routines and distillation routines for qubits with various advantages have been proposed. == Stabilizer formalism == The Clifford group consists of a set of n {\displaystyle n} -qubit operations generated by the gates {H, S, CNOT} (where H is Hadamard and S is [ 1 0 0 i ] {\displaystyle {\begin{bmatrix}1&0\\0&i\end{bmatrix}}} ) called Clifford gates. The Clifford group generates stabilizer states which can be efficiently simulated classically, as shown by the Gottesman–Knill theorem. This set of gates with a non-Clifford operation is universal for quantum computation. == Magic states == Magic states are purified from n {\displaystyle n} copies of a mixed state ρ {\displaystyle \rho } . These states are typically provided via an ancilla to the circuit. A magic state for the π / 6 {\displaystyle \pi /6} rotation operator is | M ⟩ = cos ⁡ ( β / 2 ) | 0 ⟩ + e i π 4 sin ⁡ ( β / 2 ) | 1 ⟩ {\displaystyle |M\rangle =\cos(\beta /2)|0\rangle +e^{i{\frac {\pi }{4}}}\sin(\beta /2)|1\rangle } where β = arccos ⁡ ( 1 3 ) {\displaystyle \beta =\arccos \left({\frac {1}{\sqrt {3}}}\right)} . A non-Clifford gate can be generated by combining (copies of) magic states with Clifford gates. Since a set of Clifford gates combined with a non-Clifford gate is universal for quantum computation, magic states combined with Clifford gates are also universal. == Purification algorithm for distilling |M〉 == The first magic state distillation algorithm, invented by Sergey Bravyi and Alexei Kitaev, is as follows. Input: Prepare 5 imperfect states. Output: An almost pure state having a small error probability. repeat Apply the decoding operation of the five-qubit error correcting code and measure the syndrome. If the measured syndrome is | 0000 ⟩ {\displaystyle |0000\rangle } , the distillation attempt is successful. else Get rid of the resulting state and restart the algorithm. until The states have been distilled to the desired purity.

  Read more →
• 

  Online public access catalog

  The online public access catalog (OPAC), now frequently synonymous with library catalog, is an online database of materials held by a library or group of libraries. Online catalogs have largely replaced the analog card catalogs previously used in libraries. == History == === Early online === Although a handful of experimental systems existed as early as the 1960s, the first large-scale online catalogs were developed at Ohio State University in 1975 and the Dallas Public Library in 1978. These and other early online catalog systems tended to closely reflect the card catalogs that they were intended to replace. Using a dedicated terminal or telnet client, users could search a handful of pre-coordinate indexes and browse the resulting display in much the same way they had previously navigated the card catalog. Throughout the 1980s, the number and sophistication of online catalogs grew. The first commercial systems appeared, and would by the end of the decade largely replace systems built by libraries themselves. Library catalogs began providing improved search mechanisms, including Boolean and keyword searching, as well as ancillary functions, such as the ability to place holds on items that had been checked-out. At the same time, libraries began to develop applications to automate the purchase, cataloging, and circulation of books and other library materials. These applications, collectively known as an integrated library system (ILS) or library management system, included an online catalog as the public interface to the system's inventory. Most library catalogs are closely tied to their underlying ILS system. === Stagnation and dissatisfaction === The 1990s saw a relative stagnation in the development of online catalogs. Although the earlier character-based interfaces were replaced with ones for the Web, both the design and the underlying search technology of most systems did not advance much beyond that developed in the late 1980s. At the same time, organizations outside of libraries began developing more sophisticated information retrieval systems. Web search engines like Google and popular e-commerce websites such as Amazon.com provided simpler to use (yet more powerful) systems that could provide relevancy ranked search results using probabilistic and vector-based queries. Prior to the widespread use of the Internet, the online catalog was often the first information retrieval system library users ever encountered. Now accustomed to web search engines, newer generations of library users have grown increasingly dissatisfied with the complex (and often arcane) search mechanisms of older online catalog systems. This has, in turn, led to vocal criticisms of these systems within the library community itself, and in recent years to the development of newer (often termed 'next-generation') catalogs. === Next-generation catalogs === Newer generations of library catalog systems, typically called discovery systems (or a discovery layer), are distinguished from earlier OPACs by their use of more sophisticated search technologies, including relevancy ranking and faceted search, as well as features aimed at greater user interaction and participation with the system, including tagging and reviews. These new features rely heavily on existing metadata which may be poor or inconsistent, particularly for older records. Newer catalog platforms may be independent of the organization's integrated library system (ILS), instead providing drivers that allow for the synchronization of data between the two systems. While the original online catalog interfaces were almost exclusively built by ILS vendors, libraries have increasingly sought next-generation catalogs built by enterprise search companies and open-source software projects, often led by libraries themselves. == Union catalogs == Although library catalogs typically reflect the holdings of a single library, they can also contain the holdings of a group or consortium of libraries. These systems, known as union catalogs, are usually designed to aid the borrowing of books and other materials among the member institutions via interlibrary loan. Examples of this type of catalogs include COPAC, SUNCAT, NLA Trove, and WorldCat—the last catalogs the collections of libraries worldwide. == Related systems == There are a number of systems that share much in common with library catalogs, but have traditionally been distinguished from them. Libraries utilize these systems to search for items not traditionally covered by a library catalog, although these systems are sometimes integrated into a more comprehensive discovery system. Bibliographic databases—such as Medline, ERIC, PsycINFO, Scopus, Web of Science, and many others—index journal articles and other research data. There are also a number of applications aimed at managing documents, photographs, and other digitized or born-digital items such as Digital Commons and DSpace. Particularly in academic libraries, these systems (often known as digital library systems or institutional repository systems) assist with efforts to preserve documents created by faculty and students. Electronic resource management helps librarians to track selection, acquisition, and licensing of a library's electronic information resources.

  Read more →
• 

  Semantic query

  Semantic queries allow for queries and analytics of associative and contextual nature. Semantic queries enable the retrieval of both explicitly and implicitly derived information based on syntactic, semantic and structural information contained in data. They are designed to deliver precise results (possibly the distinctive selection of one single piece of information) or to answer more fuzzy and wide open questions through pattern matching and digital reasoning. Semantic queries work on named graphs, linked data or triples. This enables the query to process the actual relationships between information and infer the answers from the network of data. This is in contrast to semantic search, which uses semantics (meaning of language constructs) in unstructured text to produce a better search result. (See natural language processing.) From a technical point of view, semantic queries are precise relational-type operations much like a database query. They work on structured data and therefore have the possibility to utilize comprehensive features like operators (e.g. >, < and =), namespaces, pattern matching, subclassing, transitive relations, semantic rules and contextual full text search. The semantic web technology stack of the W3C is offering SPARQL to formulate semantic queries in a syntax similar to SQL. Semantic queries are used in triplestores, graph databases, semantic wikis, natural language and artificial intelligence systems. == Background == Relational databases represent all relationships between data in an implicit manner only. For example, the relationships between customers and products (stored in two content-tables and connected with an additional link-table) only come into existence in a query statement (SQL in the case of relational databases) written by a developer. Writing the query demands exact knowledge of the database schema. Linked-Data represent all relationships between data in an explicit manner. In the above example, no query code needs to be written. The correct product for each customer can be fetched automatically. Whereas this simple example is trivial, the real power of linked-data comes into play when a network of information is created (customers with their geo-spatial information like city, state and country; products with their categories within sub- and super-categories). Now the system can automatically answer more complex queries and analytics that look for the connection of a particular location with a product category. The development effort for this query is omitted. Executing a semantic query is conducted by walking the network of information and finding matches (also called Data Graph Traversal). Another important aspect of semantic queries is that the type of the relationship can be used to incorporate intelligence into the system. The relationship between a customer and a product has a fundamentally different nature than the relationship between a neighbourhood and its city. The latter enables the semantic query engine to infer that a customer living in Manhattan is also living in New York City whereas other relationships might have more complicated patterns and "contextual analytics". This process is called inference or reasoning and is the ability of the software to derive new information based on given facts. == Articles == Velez, Golda (2008). "Semantics Help Wall Street Cope With Data Overload". Wall Street & Technology. wallstreetandtech.com. Zhifeng, Xiao (2009). "Spatial information semantic query based on SPARQL". In Liu, Yaolin; Tang, Xinming (eds.). International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining. Vol. 7492. SPIE. pp. 74921P. Bibcode:2009SPIE.7492E..60X. doi:10.1117/12.838556. S2CID 62191842. Aquin, Mathieu (2010). "Watson, more than a Semantic Web search engine" (PDF). Semantic Web Journal. Dworetzky, Tom (2011). "How Siri Works: iPhone's 'Brain' Comes from Natural Language Processing". International Business Times. Horwitt, Elisabeth (2011). "The semantic Web gets down to business". computerworld.com. Rodriguez, Marko (2011). "Graph Pattern Matching with Gremlin". Marko A. Rodriguez. markorodriguez.com on Graph Computing. Sequeda, Juan (2011). "SPARQL Nuts & Bolts". Cambridge Semantics. Freitas, Andre (2012). "Querying Heterogeneous Datasets on the Linked Data Web" (PDF). IEEE Internet Computing. Kauppinen, Tomi (2012). "Using the SPARQL Package in R to handle Spatial Linked Data". linkedscience.org. Lorentz, Alissa (2013). "With Big Data, Context is a Big Issue". Wired.

  Read more →
• 

  Convolutional layer

  In artificial neural networks, a convolutional layer is a type of network layer that applies a convolution operation to the input. Convolutional layers are some of the primary building blocks of convolutional neural networks (CNNs), a class of neural network most commonly applied to images, video, audio, and other data that have the property of uniform translational symmetry. The convolution operation in a convolutional layer involves sliding a small window (called a kernel or filter) across the input data and computing the dot product between the values in the kernel and the input at each position. This process creates a feature map that represents detected features in the input. == Concepts == === Kernel === Kernels, also known as filters, are small matrices of weights that are learned during the training process. Each kernel is responsible for detecting a specific feature in the input data. The size of the kernel is a hyperparameter that affects the network's behavior. === Convolution === For a 2D input x {\displaystyle x} and a 2D kernel w {\displaystyle w} , the 2D convolution operation can be expressed as: y [ i , j ] = ∑ m = 0 k h − 1 ∑ n = 0 k w − 1 x [ i + m , j + n ] ⋅ w [ m , n ] {\displaystyle y[i,j]=\sum _{m=0}^{k_{h}-1}\sum _{n=0}^{k_{w}-1}x[i+m,j+n]\cdot w[m,n]} where k h {\displaystyle k_{h}} and k w {\displaystyle k_{w}} are the height and width of the kernel, respectively. This generalizes immediately to nD convolutions. Commonly used convolutions are 1D (for audio and text), 2D (for images), and 3D (for spatial objects, and videos). === Stride === Stride determines how the kernel moves across the input data. A stride of 1 means the kernel shifts by one pixel at a time, while a larger stride (e.g., 2 or 3) results in less overlap between convolutions and produces smaller output feature maps. === Padding === Padding involves adding extra pixels around the edges of the input data. It serves two main purposes: Preserving spatial dimensions: Without padding, each convolution reduces the size of the feature map. Handling border pixels: Padding ensures that border pixels are given equal importance in the convolution process. Common padding strategies include: No padding/valid padding. This strategy typically causes the output to shrink. Same padding: Any method that ensures the output size same as input size is a same padding strategy. Full padding: Any method that ensures each input entry is convolved over for the same number of times is a full padding strategy. Common padding algorithms include: Zero padding: Add zero entries to the borders of input. Mirror/reflect/symmetric padding: Reflect the input array on the border. Circular padding: Cycle the input array back to the opposite border, like a torus. The exact numbers used in convolutions is complicated, for which we refer to (Dumoulin and Visin, 2018) for details. == Variants == === Standard === The basic form of convolution as described above, where each kernel is applied to the entire input volume. === Depthwise separable === Depthwise separable convolution separates the standard convolution into two steps: depthwise convolution and pointwise convolution. The depthwise separable convolution decomposes a single standard convolution into two convolutions: a depthwise convolution that filters each input channel independently and a pointwise convolution ( 1 × 1 {\displaystyle 1\times 1} convolution) that combines the outputs of the depthwise convolution. This factorization significantly reduces computational cost. It was first developed by Laurent Sifre during an internship at Google Brain in 2013 as an architectural variation on AlexNet to improve convergence speed and model size. === Dilated === Dilated convolution, or atrous convolution, introduces gaps between kernel elements, allowing the network to capture a larger receptive field without increasing the kernel size. === Transposed === Transposed convolution, also known as deconvolution, fractionally strided convolution, and upsampling convolution, is a convolution where the output tensor is larger than its input tensor. It's often used in encoder-decoder architectures for upsampling. It's used in image generation, semantic segmentation, and super-resolution tasks. == History == The concept of convolution in neural networks was inspired by the visual cortex in biological brains. Early work by Hubel and Wiesel in the 1960s on the cat's visual system laid the groundwork for artificial convolution networks. An early convolution neural network was developed by Kunihiko Fukushima in 1969. It had mostly hand-designed kernels inspired by convolutions in mammalian vision. In 1979 he improved it to the Neocognitron, which learns all convolutional kernels by unsupervised learning (in his terminology, "self-organized by 'learning without a teacher'"). During the 1988 to 1998 period, a series of CNN were introduced by Yann LeCun et al., ending with LeNet-5 in 1998. It was an early influential CNN architecture for handwritten digit recognition, trained on the MNIST dataset, and was used in ATM. (Olshausen & Field, 1996) discovered that simple cells in the mammalian primary visual cortex implement localized, oriented, bandpass receptive fields, which could be recreated by fitting sparse linear codes for natural scenes. This was later found to also occur in the lowest-level kernels of trained CNNs. The field saw a resurgence in the 2010s with the development of deeper architectures and the availability of large datasets and powerful GPUs. AlexNet, developed by Alex Krizhevsky et al. in 2012, was a catalytic event in modern deep learning. In that year’s ImageNet competition, the AlexNet model achieved a 16% top-five error rate, significantly outperforming the next best entry, which had a 26% error rate. The network used eight trainable layers, approximately 650,000 neurons, and around 60 million parameters, highlighting the impact of deeper architectures and GPU acceleration on image recognition performance. From the 2013 ImageNet competition, most entries adopted deep convolutional neural networks, building on the success of AlexNet. Over the following years, performance steadily improved, with the top-five error rate falling from 16% in 2012 and 12% in 2013 to below 3% by 2017, as networks grew increasingly deep.

  Read more →
• 

  Hindley–Milner type system

  A Hindley–Milner (HM) type system is a classical type system for the lambda calculus with parametric polymorphism. It is also known as Damas–Milner or Damas–Hindley–Milner. It was first described by J. Roger Hindley and later rediscovered by Robin Milner. Luis Damas contributed a close formal analysis and proof of the method in his PhD thesis. Among HM's more notable properties are its completeness and its ability to infer the most general type of a given program without programmer-supplied type annotations or other hints. Algorithm W is an efficient type inference method in practice and has been successfully applied on large code bases, although it has a high theoretical complexity. HM is preferably used for functional programming languages. It was first implemented as part of the type system of the programming language ML. Since then, HM has been extended in various ways, most notably with type class constraints like those in Haskell. == Introduction == As a type inference method, Hindley–Milner is able to deduce the types of variables, expressions and functions from programs written in an entirely untyped style. Being scope sensitive, it is not limited to deriving the types only from a small portion of source code, but rather from complete programs or modules. Being able to cope with parametric types, too, it is core to the type systems of many functional programming languages. It was first applied in this manner in the ML programming language. The origin is the type inference algorithm for the simply typed lambda calculus that was devised by Haskell Curry and Robert Feys in 1958. In 1969, J. Roger Hindley extended this work and proved that their algorithm always inferred the most general type. In 1978, Robin Milner, independently of Hindley's work, provided an equivalent algorithm, Algorithm W. In 1982, Luis Damas finally proved that Milner's algorithm is complete and extended it to support systems with polymorphic references. === Monomorphism vs. polymorphism === In the simply typed lambda calculus, types T are either atomic type constants or function types of form T → T {\displaystyle T\rightarrow T} . Such types are monomorphic. Typical examples are the types used in arithmetic values: 3 : N u m b e r a d d 3 4 : N u m b e r a d d : N u m b e r → N u m b e r → N u m b e r {\displaystyle {\begin{array}{ll}3&:{\mathtt {Number}}\\{\mathtt {add}}\ 3\ 4&:{\mathtt {Number}}\\{\mathtt {add}}&:{\mathtt {Number}}\rightarrow {\mathtt {Number}}\rightarrow {\mathtt {Number}}\end{array}}} Contrary to this, the untyped lambda calculus is neutral to typing at all, and many of its functions can be meaningfully applied to all type of arguments. The trivial example is the identity function i d ≡ λ x . x {\displaystyle {\mathtt {id}}\equiv \lambda x.x} which simply returns whatever value it is applied to. Less trivial examples include parametric types like lists. While polymorphism in general means that operations accept values of more than one type, the polymorphism used here is parametric. One finds the notation of type schemes in the literature, too, emphasizing the parametric nature of the polymorphism. Additionally, constants may be typed with (quantified) type variables. For example, the following type schemes quantify universally over α {\displaystyle \alpha } , meaning that they are true for all possible α {\displaystyle \alpha } : c o n s : ∀ α . α → L i s t α → L i s t α n i l : ∀ α . L i s t α i d : ∀ α . α → α {\displaystyle {\begin{array}{ll}{\mathtt {cons}}&:\forall \alpha .\alpha \rightarrow {\mathtt {List}}\ \alpha \rightarrow {\mathtt {List}}\ \alpha \\{\mathtt {nil}}&:\forall \alpha .{\mathtt {List}}\ \alpha \\{\mathtt {id}}&:\forall \alpha .\alpha \rightarrow \alpha \end{array}}} Polymorphic types can become monomorphic by consistent substitution of their variables. Examples of monomorphic instances are: i d ′ : S t r i n g → S t r i n g n i l ′ : L i s t N u m b e r {\displaystyle {\begin{array}{ll}{\mathtt {id}}'&:{\mathtt {String}}\rightarrow {\mathtt {String}}\\{\mathtt {nil}}'&:{\mathtt {List}}\ {\mathtt {Number}}\end{array}}} More generally, types are polymorphic when they contain type variables, while types without them are monomorphic. Contrary to the type systems used for example in Pascal (1970) or C (1972), which only support monomorphic types, HM is designed with emphasis on parametric polymorphism. The successors of the languages mentioned, like C++ (1985), focused on different types of polymorphism, namely subtyping in connection with object-oriented programming and overloading. While subtyping is incompatible with HM, a variant of systematic overloading is available in the HM-based type system of Haskell. === Let-polymorphism === When extending the type inference for the simply-typed lambda calculus towards polymorphism, one has to decide whether assigning a polymorphic type not only as type of an expression, but also as the type of a λ-bound variable is admissible. This would allow the generic identity type to be assigned to the variable 'id' in: (λ id . ... (id 3) ... (id "text") ... ) (λ x . x) Allowing this gives rise to the polymorphic lambda calculus; however, type inference in this system is not decidable. Instead, HM distinguishes variables that are immediately bound to an expression from more general λ-bound variables, calling the former let-bound variables, and allows polymorphic types to be assigned only to these. This leads to let-polymorphism where the above example takes the form let id = λ x . x in ... (id 3) ... (id "text") ... which can be typed with a polymorphic type for 'id'. As indicated, the expression syntax is extended to make the let-bound variables explicit, and by restricting the type system to allow only let-bound variable to have polymorphic types, while the parameters in lambda-abstractions must get a monomorphic type, type inference becomes decidable. == Overview == The remainder of this article proceeds as follows: The HM type system is defined. This is done by describing a deduction system that makes precise what expressions have what type, if any. From there, it works towards an implementation of the type inference method. After introducing a syntax-driven variant of the above deductive system, it sketches an efficient implementation (algorithm J), appealing mostly to the reader's metalogical intuition. Because it remains open whether algorithm J indeed realises the initial deduction system, a less efficient implementation (algorithm W), is introduced and its use in a proof is hinted. Finally, further topics related to the algorithm are discussed. The same description of the deduction system is used throughout, even for the two algorithms, to make the various forms in which the HM method is presented directly comparable. == The Hindley–Milner type system == The type system can be formally described by syntax rules that fix a language for the expressions, types, etc. The presentation here of such a syntax is not too formal, in that it is written down not to study the surface grammar, but rather the depth grammar, and leaves some syntactical details open. This form of presentation is usual. Building on this, typing rules are used to define how expressions and types are related. As before, the form used is a bit liberal. === Syntax === The expressions to be typed are exactly those of the lambda calculus extended with a let-expression as shown in the adjacent table. Parentheses can be used to disambiguate an expression. The application is left-binding and binds stronger than abstraction or the let-in construct. Types are syntactically split into two groups, monotypes and polytypes. ==== Monotypes ==== Monotypes always designate a particular type. Monotypes τ {\displaystyle \tau } are syntactically represented as terms. Examples of monotypes include type constants like i n t {\displaystyle {\mathtt {int}}} or s t r i n g {\displaystyle {\mathtt {string}}} , and parametric types like M a p ( S e t s t r i n g ) i n t {\displaystyle {\mathtt {Map\ (Set\ string)\ int}}} . The latter types are examples of applications of type functions, for example, from the set { M a p 2 , S e t 1 , s t r i n g 0 , i n t 0 , → 2 } {\displaystyle \{{\mathtt {Map^{2},\ Set^{1},\ string^{0},\ int^{0}}},\ \rightarrow ^{2}\}} , where the superscript indicates the number of type parameters. The complete set of type functions C {\displaystyle C} is arbitrary in HM, except that it must contain at least → 2 {\displaystyle \rightarrow ^{2}} , the type of functions. It is often written in infix notation for convenience. For example, a function mapping integers to strings has type i n t → s t r i n g {\displaystyle {\mathtt {int}}\rightarrow {\mathtt {string}}} . Again, parentheses can be used to disambiguate a type expression. The application binds stronger than the infix arrow, which is right-binding. Type variables are admitted as monotypes. Monotypes are not to be confused with monomorphic types, which exc

  Read more →
• 

  AVT Statistical filtering algorithm

  AVT Statistical filtering algorithm is an approach to improving quality of raw data collected from various sources. It is most effective in cases when there is inband noise present. In those cases AVT is better at filtering data then, band-pass filter or any digital filtering based on variation of. Conventional filtering is useful when signal/data has different frequency than noise and signal/data is separated/filtered by frequency discrimination of noise. Frequency discrimination filtering is done using Low Pass, High Pass and Band Pass filtering which refers to relative frequency filtering criteria target for such configuration. Those filters are created using passive and active components and sometimes are implemented using software algorithms based on Fast Fourier transform (FFT). AVT filtering is implemented in software and its inner working is based on statistical analysis of raw data. When signal frequency/(useful data distribution frequency) coincides with noise frequency/(noisy data distribution frequency) we have inband noise. In this situations frequency discrimination filtering does not work since the noise and useful signal are indistinguishable and where AVT excels. To achieve filtering in such conditions there are several methods/algorithms available which are briefly described below. == Averaging algorithm == Collect n samples of data Calculate average value of collected data Present/record result as actual data == Median algorithm == Collect n samples of data Sort the data in ascending or descending order. Note that order does not matter Select the data that happen to be in n/2 position and present/record it as final result representing data sample == AVT algorithm == AVT algorithm stands for Antonyan Vardan Transform and its implementation explained below. Collect n samples of data Calculate the standard deviation and average value Drop any data that is greater or less than average ± one standard deviation Calculate average value of remaining data Present/record result as actual value representing data sample This algorithm is based on amplitude discrimination and can easily reject any noise that is not like actual signal, otherwise statistically different than 1 standard deviation of the signal. Note that this type of filtering can be used in situations where the actual environmental noise is not known in advance. Notice that it is preferable to use the median in above steps than average. Originally the AVT algorithm used average value to compare it with results of median on the data window. == Filtering algorithms comparison == Using a system that has signal value of 1 and has noise added at 0.1% and 1% levels will simplify quantification of algorithm performance. The R script is used to create pseudo random noise added to signal and analyze the results of filtering using several algorithms. Please refer to "Reduce Inband Noise with the AVT Algorithm" article for details. This graphs show that AVT algorithm provides best results compared with Median and Averaging algorithms while using data sample size of 32, 64 and 128 values. Note that this graph was created by analyzing random data array of 10000 values. Sample of this data is graphically represented below. From this graph it is apparent that AVT outperforms other filtering algorithms by providing 5% to 10% more accurate data when analyzing same datasets. Considering random nature of noise used in this numerical experiment that borderlines worst case situation where actual signal level is below ambient noise the precision improvements of processing data with AVT algorithm are significant. == AVT algorithm variations == === Cascaded AVT === In some situations better results can be obtained by cascading several stages of AVT filtering. This will produce singular constant value which can be used for equipment that has known stable characteristics like thermometers, thermistors and other slow acting sensors. === Reverse AVT === Collect n samples of data Calculate the standard deviation and average value Drop any data that is within one standard deviation ± average band Calculate average value of remaining data Present/record result as actual data This is useful for detecting minute signals that are close to background noise level. == Possible applications and uses == Use to filter data that is near or below noise level Used in planet detection to filter out raw data from the Kepler space telescope Filter out noise from sound sources where all other filtering methods (Low-pass filter, High-pass filter, Band-pass filter, Digital filter) fail. Pre-process scientific data for data analysis (Smoothness) before plotting see (Plot (graphics)) Used in SETI (Search for extraterrestrial intelligence) for detecting/distinguishing extraterrestrial signals from cosmic background Use AVT as image filtering algorithm to detect altered images. This image of Jupiter generated from this program, detecting alterations in original picture that was modified to be visually appealing by applying filters. Another version of this comparison is the Reverse AVT filter applied to the same original Jupiter Image, where we only see that altered portion as Noise that was eliminated by AVT algorithm. Use AVT as image filtering algorithm to estimate data density from images. Picture of Pillars of Creation Nebula shows data density in filtered images from Hubble and Webb. Note that image on the left has big patches of missing data marked with simpler color patterns.

  Read more →
• 

  Collision problem

  The r-to-1 collision problem is an important theoretical problem in complexity theory, quantum computing, and computational mathematics. The collision problem most often refers to the 2-to-1 version: given n {\displaystyle n} even and a function f : { 1 , … , n } → { 1 , … , n } {\displaystyle f:\,\{1,\ldots ,n\}\rightarrow \{1,\ldots ,n\}} , we are promised that f is either 1-to-1 or 2-to-1. We are only allowed to make queries about the value of f ( i ) {\displaystyle f(i)} for any i ∈ { 1 , … , n } {\displaystyle i\in \{1,\ldots ,n\}} . The problem then asks how many such queries we need to make to determine with certainty whether f is 1-to-1 or 2-to-1. == Classical solutions == === Deterministic === Solving the 2-to-1 version deterministically requires n 2 + 1 {\textstyle {\frac {n}{2}}+1} queries, and in general distinguishing r-to-1 functions from 1-to-1 functions requires n r + 1 {\textstyle {\frac {n}{r}}+1} queries. This is a straightforward application of the pigeonhole principle: if a function is r-to-1, then after n r + 1 {\textstyle {\frac {n}{r}}+1} queries we are guaranteed to have found a collision. If a function is 1-to-1, then no collision exists. Thus, n r + 1 {\textstyle {\frac {n}{r}}+1} queries suffice. If we are unlucky, then the first n / r {\displaystyle n/r} queries could return distinct answers, so n r + 1 {\textstyle {\frac {n}{r}}+1} queries is also necessary. === Randomized === If we allow randomness, the problem is easier. By the birthday paradox, if we choose (distinct) queries at random, then with high probability we find a collision in any fixed 2-to-1 function after Θ ( n ) {\displaystyle \Theta ({\sqrt {n}})} queries. == Quantum solution == The BHT algorithm, which uses Grover's algorithm, solves this problem optimally by only making O ( n 1 / 3 ) {\displaystyle O(n^{1/3})} queries to f. The matching lower bound of Ω ( n 1 / 3 ) {\displaystyle \Omega (n^{1/3})} was proved by Aaronson and Shi using the polynomial method.

  Read more →
• 

  Swizzling (computer graphics)

  In computer graphics, swizzles are a class of operations that transform vectors by rearranging components. Swizzles can also project from a vector of one dimensionality to a vector of another dimensionality, such as taking a three-dimensional vector and creating a two-dimensional or five-dimensional vector using components from the original vector. For example, if A = {1,2,3,4}, where the components are x, y, z, and w respectively, one could compute B = A.wwxy, whereupon B would equal {4,4,1,2}. Additionally, one could create a two-dimensional vector with A.wx or a five-dimensional vector with A.xyzwx. Combining vectors and swizzling can be employed in various ways. This is common in GPGPU applications. In terms of linear algebra, this is equivalent to multiplying by a matrix whose rows are standard basis vectors. If A = ( 1 , 2 , 3 , 4 ) T {\displaystyle A=(1,2,3,4)^{T}} , then swizzling A {\displaystyle A} as above looks like A . w w x y = [ 0 0 0 1 0 0 0 1 1 0 0 0 0 1 0 0 ] [ 1 2 3 4 ] = [ 4 4 1 2 ] . {\displaystyle A.\!wwxy={\begin{bmatrix}0&0&0&1\\0&0&0&1\\1&0&0&0\\0&1&0&0\end{bmatrix}}{\begin{bmatrix}1\\2\\3\\4\end{bmatrix}}={\begin{bmatrix}4\\4\\1\\2\end{bmatrix}}.}

  Read more →
• 

  Interviewer effect

  The interviewer effect (also called interviewer variance or interviewer error) is the distortion of response to an interviewer-administered data collection effort which results from differential reactions to the social style and personality of interviewers or to their presentation of particular questions. The use of fixed-wording questions is one method of reducing interviewer bias. Anthropological research and case-studies are also affected by the problem, which is exacerbated by the self-fulfilling prophecy, when the researcher is also the interviewer it is also any effect on data gathered from interviewing people that is caused by the behavior or characteristics (real or perceived) of the interviewer. Interviewer effects can also be associated with the characteristics of the interviewer, such as race. Whether black respondents are interviewed by white interviewers or black interviewers has a strong impact on their responses to both attitude questions and behavioral ones. In the latter case, for example, if black respondents are interviewed by black interviewers in pre-election surveys, they are more likely to actually vote in the upcoming election than if they are interviewed by white interviewers. Furthermore, the race of the interviewer can also affect answers to factual questions that might take the form of a test of how informed the respondent is. Black respondents in a survey of political knowledge, for example, get fewer correct answers to factual questions about politics when interviewed by white interviewers than when interviewed by black interviewers. This is consistent with the research literature on stereotype threat, which finds diminished test performance of potentially stigmatised groups when the interviewer or test supervisor is from a perceived higher status group. Interviewer effects can be mitigated somewhat by randomly assigning subjects to different interviewers, or by using tools such as computer-assisted telephone interviewing (CATI).

  Read more →
• 

  Emotion Markup Language

  An Emotion Markup Language (EML or EmotionML) has first been defined by the W3C Emotion Incubator Group (EmoXG) as a general-purpose emotion annotation and representation language, which should be usable in a large variety of technological contexts where emotions need to be represented. Emotion-oriented computing (or "affective computing") is gaining importance as interactive technological systems become more sophisticated. Representing the emotional states of a user or the emotional states to be simulated by a user interface requires a suitable representation format; in this case a markup language is used. EmotionML version 1.0 was published by the group in May 2014. == Example == Here is an example of an EmotionML document describing emotions expressed in a video recording of the interaction between a teacher, Alice, and a student, Bob. == History == In 2006, a first W3C Incubator Group, the Emotion Incubator Group (EmoXG), was set up "to investigate a language to represent the emotional states of users and the emotional states simulated by user interfaces" with the final Report published on 10 July 2007. In 2007, the Emotion Markup Language Incubator Group (EmotionML XG) was set up as a follow-up to the Emotion Incubator Group, "to propose a specification draft for an Emotion Markup Language, to document it in a way accessible to non-experts, and to illustrate its use in conjunction with a number of existing markups." The final report of the Emotion Markup Language Incubator Group, Elements of an EmotionML 1.0, was published on 20 November 2008. The work then was continued in 2009 in the frame of the W3C's Multimodal Interaction Activity, with the First Public Working Draft of "Emotion Markup Language (EmotionML) 1.0" being published on 29 October 2009. The Last Call Working Draft of "Emotion Markup Language 1.0", was published on 7 April 2011. The Last Call Working Draft addressed all open issues that arose from feedback of the community on the First Call Working Draft as well as results of a workshop held in Paris in October 2010. Along with the Last Call Working Draft, a list of vocabularies for EmotionML has been published to aid developers using common vocabularies for annotating or representing emotions. Annual draft updates were published until the 1.0 version was finished in 2014. == Reasons for defining an emotion markup language == A standard for an emotion markup language would be useful for the following purposes: To enhance computer-mediated human-human or human-machine communication. Emotions are a basic part of human communication and should therefore be taken into account, e.g. in emotional Chat systems or emphatic voice boxes. This involves specification, analysis and display of emotion related states. To enhance systems' processing efficiency. Emotion and intelligence are strongly interconnected. The modeling of human emotions in computer processing can help to build more efficient systems, e.g. using emotional models for time-critical decision enforcement. To allow the analysis of non-verbal behavior, emotion, mental states that can be provided using web services to enable data collection, analysis, and reporting. Concrete examples of existing technology that could apply EmotionML include: Opinion mining / sentiment analysis in Web 2.0, to automatically track customer's attitude regarding a product across blogs; Affective monitoring, such as ambient assisted living applications, fear detection for surveillance purposes, or using wearable sensors to test customer satisfaction; Wellness technologies that provide assistance according to a person's emotional state with the goal to improve the person's well-being; Character design and control for games and virtual worlds; Building web services to capture, analysis, and report data of non-verbal behavior, emotion and mental states of an individual or group across the internet using standard web technologies such as HTML5 and JSON. Social robots, such as guide robots engaging with visitors; Expressive speech synthesis, generating synthetic speech with different emotions, such as happy or sad, friendly or apologetic; expressive synthetic speech would for example make more information available to blind and partially sighted people, and enrich their experience of the content; Emotion recognition (e.g., for spotting angry customers in speech dialog systems, to improve computer games or e-Learning applications); Support for people with disabilities, such as educational programs for people with autism. EmotionML can be used to make the emotional intent of content explicit. This would enable people with learning disabilities (such as Asperger syndrome) to realise the emotional context of the content; EmotionML can be used for media transcripts and captions. Where emotions are marked up to help deaf or hearing impaired people who cannot hear the soundtrack, more information is made available to enrich their experience of the content. The Emotion Incubator Group has listed 39 individual use cases for an Emotion markup language. A standardised way to mark up the data needed by such "emotion-oriented systems" has the potential to boost development primarily because data that was annotated in a standardised way can be interchanged between systems more easily, thereby simplifying a market for emotional databases, and the standard can be used to ease a market of providers for sub-modules of emotion processing systems, e.g. a web service for the recognition of emotion from text, speech or multi-modal input. == The challenge of defining a generally usable emotion markup language == Any attempt to standardize the description of emotions using a finite set of fixed descriptors is doomed to failure, as there is no consensus on the number of relevant emotions, on the names that should be given to them or how else best to describe them. For example, the difference between ":)" and "(:" is small, but using a standardized markup it would make one invalid. Even more basically, the list of emotion-related states that should be distinguished varies depending on the application domain and the aspect of emotions to be focused. Basically, the vocabulary needed depends on the context of use. On the other hand, the basic structure of concepts is less controversial: it is generally agreed that emotions involve triggers, appraisals, feelings, expressive behavior including physiological changes, and action tendencies; emotions in their entirety can be described in terms of categories or a small number of dimensions; emotions have an intensity, and so on. For details, see the Scientific Descriptions of Emotions in the Final Report of the Emotion Incubator Group. Given this lack of agreement on descriptors in the field, the only practical way of defining an emotion markup language is the definition of possible structural elements and to allow users to "plug in" vocabularies that they consider appropriate for their work. An additional challenge lies in the aim to provide a markup language that is generally usable. The requirements that arise from different use cases are rather different. Whereas manual annotation tends to require all the fine-grained distinctions considered in the scientific literature, automatic recognition systems can usually distinguish only a very small number of different states and affective avatars need yet another level of detail for expressing emotions in an appropriate way. For the reasons outlined here, it is clear that there is an inevitable tension between flexibility and interoperability, which need to be weighed in the formulation of an EmotionML. The guiding principle in the following specification has been to provide a choice only where it is needed, and to propose reasonable default options for every choice. == Applications and web services benefiting from an emotion markup language == There are a range of existing projects and applications to which an emotion markup language will enable the building of webservices to measure capture data of individuals non-verbal behavior, mental states, and emotions and allowing results to be reported and rendered in a standardized format using standard web technologies such as JSON and HTML5. One such project is measuring affect data across the Internet using EyesWeb.

  Read more →
• 

  National Data Repository

  A National Data Repository (NDR) is a data bank that seeks to preserve and promote a country's natural resources data, particularly data related to the petroleum exploration and production (E&P) sector. A National Data Repository is normally established by an entity that governs, controls and supports the exchange, capture, transference and distribution of E&P information, with the final target to provide the State with the tools and information to assure the growth, govern-ability, control, independence and sovereignty of the industry. The two fundamental reasons for a country to establish an NDR are to preserve data generated inside the country by the industry, and to promote investments in the country by utilizing data to reduce the exploration, production, and transportation business risks. Countries take different approaches towards preserving and promoting their natural resources data. The approach varies according to a country's natural resources policies, level of openness, and its attitude towards foreign investment. == Data types == NDRs store a vast array of data related to a country's natural resources. This includes wells, well log data, well reports, core samples, seismic surveys, post-stack seismic, field data/tapes, seismic (acquisition/processing) reports, production data, geological maps and reports, license data and geological models. == Funding models == Some NDRs are financed entirely by a country's government. Others are industry-funded. Still some are hybrid systems, funded in part by industry and government. NDRs typically charge fees for data requests and for data loading. The cost differs significantly between countries. In some cases an annual membership is charged to oil companies to store and access the data in the NDR. == Standards body == Energistics is the global energy standards resource center for the upstream oil and gas industry. Energistics National Data Repository Work Group: The standards body is Energistics. === Energistics-standards-directory === Global regulators of upstream oil and natural gas information, including seismic, drilling, production and reservoir data, formed the National Data Repository (NDR) Work Group in 2008 to collaborate on the development of data management standards and to assist emerging nations with hydrocarbon reserves to better collect, maintain and deliver oil and gas data to the public and to the industry. Ten countries, led by the Netherlands, Norway and the United Kingdom, formed NDR to share best practices and to formalize the development and deployment of data management standards for regulatory agencies. The other countries involved in the NDR Work Group's formation are Australia, Canada, India, Kenya, New Zealand, South Africa and the United States. Annual NDR Conference: Approximately every 18 months Energistics organizes a National Data Repository Conference. The purpose is to provide government and regulatory agencies from around the world an opportunity to attend a series of workshops dedicated to developing data exchange standards, improving communications with the oil and gas industry and learning data management techniques for natural resources information. === Society of Exploration Geophysicists and The International Oil and Gas Producers Association === The SEG is the custodian of the SEG standards which are used for the exchange, retention and release of seismic data. They are commonly used by National Data Repositories with the SEGD and SEGY being the field and processed exchange standards respectively. == NDRs around the world == Click here to see a map of the NDRs around the world

  Read more →
• 

  Pulse-coupled networks

  Pulse-coupled networks or pulse-coupled neural networks (PCNNs) are neural models proposed by modeling a cat's visual cortex, and developed for high-performance biomimetic image processing. In 1989, Eckhorn introduced a neural model to emulate the mechanism of cat's visual cortex. The Eckhorn model provided a simple and effective tool for studying small mammal’s visual cortex, and was soon recognized as having significant application potential in image processing. In 1994, Johnson adapted the Eckhorn model to an image processing algorithm, calling this algorithm a pulse-coupled neural network. The basic property of the Eckhorn's linking-field model (LFM) is the coupling term. LFM is a modulation of the primary input by a biased offset factor driven by the linking input. These drive a threshold variable that decays from an initial high value. When the threshold drops below zero it is reset to a high value and the process starts over. This is different than the standard integrate-and-fire neural model, which accumulates the input until it passes an upper limit and effectively "shorts out" to cause the pulse. LFM uses this difference to sustain pulse bursts, something the standard model does not do on a single neuron level. It is valuable to understand, however, that a detailed analysis of the standard model must include a shunting term, due to the floating voltages level in the dendritic compartment(s), and in turn this causes an elegant multiple modulation effect that enables a true higher-order network (HON). A PCNN is a two-dimensional neural network. Each neuron in the network corresponds to one pixel in an input image, receiving its corresponding pixel's color information (e.g. intensity) as an external stimulus. Each neuron also connects with its neighboring neurons, receiving local stimuli from them. The external and local stimuli are combined in an internal activation system, which accumulates the stimuli until it exceeds a dynamic threshold, resulting in a pulse output. Through iterative computation, PCNN neurons produce temporal series of pulse outputs. The temporal series of pulse outputs contain information of input images and can be used for various image processing applications, such as image segmentation and feature generation. Compared with conventional image processing means, PCNNs have several significant merits, including robustness against noise, independence of geometric variations in input patterns, capability of bridging minor intensity variations in input patterns, etc. A simplified PCNN called a spiking cortical model was developed in 2009. == Applications == PCNNs are useful for image processing, as discussed in a book by Thomas Lindblad and Jason M. Kinser. PCNNs have been used in a variety of image processing applications, including: image segmentation, pattern recognition, feature generation, face extraction, motion detection, region growing, image denoising and image enhancement Multidimensional pulse image processing of chemical structure data using PCNN has been discussed by Kinser, et al. They have also been applied to an all pairs shortest path problem.

  Read more →
• 

  AI: When a Robot Writes a Play

  AI: When a Robot Writes a Play (in Czech: AI: Když robot píše hru) is a 2021 experimental theatre play, where 90% of its script was automatically generated by artificial intelligence (the GPT-2 language model). The play is in Czech language, but an English version of the script also exists. == Creation == The play is the first result of the THEaiTRE research project, aiming to commemorate the centenary of the R.U.R. play by Karel Čapek by investigating to what extent artificial intelligence could be used to create theatre play scripts. The script of the play was created using the THEaiTRobot tool, based on the GPT-2 language model. First, the play dramaturge, David Košťák, described the initial setting of each scene in a few sentences, and wrote the first line for each character. Next, THEaiTRobot suggested a continuation of the script, which the dramaturge could use, reject, or use part of it and let the tool generate a new continuation. Another option was to manually insert another line or a scenic remark. The script was generated in English and was automatically translated to Czech by the state-of-the-art CUBBITT machine translation tool. The resulting script was then further post-edited by the dramaturge. The resulting script was made freely available for non-commercial use both in English and in Czech, with marked manually inserted texts and manual edits. The analysis shows that 90% of the English script is automatically generated, with 10% manually written or manually post-edited. In the Czech script, a larger amount of edits were made, but the analysis claims that these additional edits are corrections of errors of the automated translation and stylistic corrections which do not change the meaning of the lines as represented by the English script, but rather bring the Czech script closer to the English one. == Characters == The play contains 9 characters. The Robot appears in all the scenes, while each of the other characters appears in only one scene. Robot – The lead character, a male humanoid robot. Master – An old man, the creator of the Robot. Boy – A schoolboy. Masseuse – A sex worker in a brothel. Stranger – An engineer. Man. Psychologist. Administrator – A female clerk at an employment agency. Actress – A film actress and a model in a robot-like costume. == Plot == The play is composed of 8 scenes. It tells the story of a humanoid robot, who encounters 8 other characters and engages into various typically human situations and activities, related to death, love, sex, violence, etc. The individual scenes are not tightly linked, but there are some linking points, such as the central character of the robot or some repeated and developing themes, such as the robot's search for love. The scenes often contain some absurd turns and it is often hard to find sense in them. It is therefore a very complicated piece interpretationally, requiring the director and the actors to invest a lot of effort and creativity in finding a meaningful interpretation which would not deviate from the script. In the interpretation by Švanda theatre, who premiered the play and who also participated on the creation of the script, the scenes typically contain non-verbally expressed content which can add a lot to the meaning of the scene compared to what is contained in the actual script (as the script only contains the lines said by the characters). === Scene 1: Death === The play opens by the Robot parting with his dying Master. The Master gives the Robot several last lessons and talks with him about death, soul, and love. === Scene 2: Sense of Humour === In the second scene, the Robot meets a sad and angry Boy, who complains that he wants to go to school, that his girlfriend is crazy, that he wants to buy a car, etc. The Robot tries to help the Boy by giving him advice, but the Boy's reactions are quite negative and irritated. The Boy then repeatedly asks the Robot to tell him a joke; the Robot keeps refusing, but ultimately tells the following joke: When you are dead. When your children are dead. When your grandchildren are dead, I will be still alive. === Scene 3: Nightclub === The Robot wants to feel pleasure, so he goes to a "night club" (a brothel), where he meets a "Masseuse" (a prostitute). The Robot is initially "a bit cold", but eventually manages to enjoy the experience and falls in love with the Masseuse. In the Švanda theatre performance, the Robot and the Masseuse seem to have a sort of virtual sex without touching each other, reminiscent of the sex scene in Demolition Man. === Scene 4: Fear of the Dark === It is the night. The Robot is standing under a lamp, unable to move away from the light as he finds that he is afraid of the dark. He meets a Stranger, an engineer who tells him that robots don't have feelings and that people cannot be trusted, and keeps hurting him. In the Švanda theatre performance, the Man repeatedly zaps the Robot with some kind of electric pulse. === Scene 5: Killer Robot === A Man approaches the Robot and repeatedly asks him to kill him. Instead, the Robot sticks a finger into the Man's anus, which leads to an argument between the Man and the Robot. === Scene 6: Burn Out === The Robot meets a Psychologist, who keeps asking him lots of questions regarding his life, burnout feeling, love, relationships, and emotions. They also talk about the Robot using a device called emotion machine which helps him to get rid of stress. === Scene 7: Search for Job === The Robot comes to an employment agency. He meets an Administrator and asks her to help him find a job. He expresses the wish to become an actor, and talks about his experience as a clown. He reveals his name to be Troy McClure, which is a character from The Simpsons who is an actor. In the Švanda theatre performance, the Administrator starts to seduce the Robot once his name is revealed, which he keeps ignoring; the Administrator then becomes irritated. === Scene 8: Love at First Sight === The Robot meets a human Actress in a robotic costume and falls in love with her immediately. The Actress is first reluctant, but the Robot manages to seduce her and she also falls in love with him. The Robot tells her about a binary world, in which he lives and where he will also take her. Ultimately, the Actress agrees, and the whole play concludes by the Robot and the Actress promising each to other to always be together. In the Švanda theatre performance, the Robot does not have a physical body in this scene, we can only hear his voice and see a pulsating light (based on the line in the script where the Robot says: "I have no body. So I don't need to wear clothes. You can't see me, you only hear me."), and the Actress eventually also agrees to lose her physical body so that she can be with the Robot forever. == Theatrical performances == The play premiered on 26 February 2021 in Švanda Theatre in Prague, Czech Republic, directed by Daniel Hrbek. Due to the COVID-19 pandemic, the play was not played in front of a live audience, but it was broadcast online, in Czech language with English subtitles. The play was followed by a panel discussion by the project members and experts on artificial intelligence. The premiere was viewed by 13,498 spectators worldwide. A short trailer of the premiere is available on YouTube. In 2021, after the opening of the theatres in the Czech Republic to spectators, the play can be viewed at Švanda Theatre. The performance takes approximately 60 minutes, and is followed by a discussion of the creators with the audience. The derniere is planned for 4 February 2023. == Reception == The play received a number of reviews, both in its country of origin as well as internationally. It is praised as first of its kind, although some reviewers note the similarity to previous works, such as the musical Beyond the Fence, the play Lifestyle of the Richard and Family, or the short movie Sunspring; however, these works used less advanced technology, and either were very short (Sunspring) or necessitated a larger amount of human interventions. The reviewers note that the script is far from perfect, with many inconsistencies and nonsensical parts, and conclude that the technology is definitely not yet ready to replace human authors; however, some find some parts of the script frighteningly human-like. The amount of human intervention is a somewhat controversial topic, with some reviewers finding the human influence too large (especially in interpreting the script and putting the play on scene), while others feel that a greater amount of human intervention would have been favorable as this could greatly improve the quality of the play. The reviews also frequently comment on the amount of sex, violence and strong language in the play; this can be attributed to the method used for creating the script, where the GPT-2 language model reflects topics and language common in the human-written articles on the internet that were used to train the model. Furthermore, some r

  Read more →
• 

  Kullback–Leibler Upper Confidence Bound

  In multi-armed bandit problems, KL-UCB (for Kullback–Leibler Upper Confidence Bound) is a UCB-type algorithm that is asymptotically optimal, in the sense that its regret matches the problem-dependent Lai-Robbins lower bound. == Multi-armed bandit problem == The Multi-armed bandit problem is a sequential game where one player has to choose at each turn between K {\displaystyle K} actions (arms). Behind every arm a {\displaystyle a} there is an unknown distribution ν a {\displaystyle \nu _{a}} that lies in a set D {\displaystyle {\mathcal {D}}} known by the player (for example, D {\displaystyle {\mathcal {D}}} can be the set of Gaussian distributions or Bernoulli distributions). At each turn t {\displaystyle t} the player chooses (pulls) an arm a t {\displaystyle a_{t}} , he then gets an observation X t {\displaystyle X_{t}} of the distribution ν a t {\displaystyle \nu _{a_{t}}} . === Regret minimization === The goal is to minimize the regret at time T {\displaystyle T} that is defined as R T := ∑ a = 1 K Δ a E [ N a ( T ) ] {\displaystyle R_{T}:=\sum _{a=1}^{K}\Delta _{a}\mathbb {E} [N_{a}(T)]} where μ a := E [ ν a ] {\displaystyle \mu _{a}:=\mathbb {E} [\nu _{a}]} is the mean of arm a {\displaystyle a} μ ∗ := max a μ a {\displaystyle \mu ^{}:=\max _{a}\mu _{a}} is the highest mean Δ a := μ ∗ − μ a {\displaystyle \Delta _{a}:=\mu ^{}-\mu _{a}} N a ( t ) {\displaystyle N_{a}(t)} is the number of pulls of arm a {\displaystyle a} up to turn t {\displaystyle t} The player has to find an algorithm that chooses at each turn t {\displaystyle t} which arm to pull based on the previous actions and observations ( a s , X s ) s < t {\displaystyle (a_{s},X_{s})_{s μ } {\displaystyle {\mathcal {K}}_{inf}(\nu ,\mu ,{\mathcal {D}}):=\inf \left\{\mathrm {KL} (\nu ,{\tilde {\nu }})\ |\ {\tilde {\nu }}\in {\mathcal {D}},\ \mathbb {E} [{\tilde {\nu }}]>\mu \right\}} K L {\displaystyle \mathrm {KL} } is the Kullback–Leibler divergence ν ^ a ( t ) {\displaystyle {\hat {\nu }}_{a}(t)} is the empirical distribution of arm a {\displaystyle a} at turn t {\displaystyle t} δ t {\displaystyle \delta _{t}} is a well-chosen sequence of positive numbers, often equal to ln ⁡ t + c ln ⁡ ln ⁡ t {\displaystyle \ln t+c\ln \ln t} with c > 0 {\displaystyle c>0} . Then we choose the arm a t {\displaystyle a_{t}} with the highest index: a t := arg ⁡ max a U a ( t ) {\displaystyle a_{t}:=\arg \max _{a}U_{a}(t)} We note that the algorithm does not require knowledge of T {\displaystyle T} . === Example === In the special case of Gaussian distribution with fixed variance σ 2 {\displaystyle \sigma ^{2}} , we have: U a ( t ) = μ ^ a ( t ) + 2 σ 2 δ t N a ( t ) {\displaystyle U_{a}(t)={\hat {\mu }}_{a}(t)+{\sqrt {\frac {2\sigma ^{2}\delta _{t}}{N_{a}(t)}}}} with μ ^ a ( t ) {\displaystyle {\hat {\mu }}_{a}(t)} being the empirical mean of arm a {\displaystyle a} at turn t {\displaystyle t} . === Pseudocode === The player gets the set D for each arm i do: n[i] ← 1; nu[i] ← None; d ← ln(K) for t from 1 to K do: select arm t observe reward r n[t] ← n[t] + 1 nu[t] ← update empirical distribution for t from K+1 to T do: for each arm i do: index[i] ← compute_index(n[i], nu[i], D, d) select arm a with highest index[a] observe reward r n[a] ← n[a] + 1 nu[a] ← update empirical distribution d ← ln(t+1) == Theoretical results == In the multi-armed bandit problem we have the Lai–Robbins asymptotic lower bound on regret. The algorithm KL-UCB matches this lower bound for one-dimensional exponential families with δ t := ln ⁡ t + 3 ln ⁡ ln ⁡ t {\displaystyle \delta _{t}:=\ln t+3\ln \ln t} and for distributions bounded in [ 0 , 1 ] {\displaystyle [0,1]} with δ t := ln ⁡ t + ln ⁡ ln ⁡ t {\displaystyle \delta _{t}:=\ln t+\ln \ln t} . === Lai–Robbins lower bound === In 1985 Lai and Robbins proved an asymptotic, problem-dependent lower bound on regret. It states that for every consistent algorithm on the set D {\displaystyle {\mathcal {D}}} — that is, an algorithm for which, for every ( ν 1 , … , ν K ) ∈ D K {\displaystyle (\nu _{1},\dots ,\nu _{K})\in {\mathcal {D}}^{K}} , the regret R T {\displaystyle R_{T}} is subpolynomial (i.e. R T = o T → + ∞ ( T α ) {\displaystyle R_{T}=o_{T\to +\infty }(T^{\alpha })} for all α > 0 {\displaystyle \alpha >0} ) — we have: R T ≥ ( ∑ a : μ a < μ ∗ Δ a K inf ( ν a , μ ∗ , D ) ) ln ⁡ T + o T → + ∞ ( ln ⁡ T ) . {\displaystyle R_{T}\geq \left(\sum _{a:\mu _{a}<\mu ^{}}{\frac {\Delta _{a}}{{\mathcal {K}}_{\inf }(\nu _{a},\mu ^{},{\mathcal {D}})}}\right)\ln T+o_{T\to +\infty }(\ln T).} This bound is asymptotic (as T → + ∞ {\displaystyle T\to +\infty } ) and gives a first-order lower bound of order ln ⁡ T {\displaystyle \ln T} with the optimal constant in front of it. === Regret bound for KL-UCB === The algorithm matches the Lai–Robbins lower bound for one-dimensional exponential-family distributions and for distributions bounded in [ 0 , 1 ] {\displaystyle [0,1]} . ==== One-dimensional exponential family ==== For D {\displaystyle {\mathcal {D}}} being the set of one-dimensional exponential families, with δ t := ln ⁡ t + 3 ln ⁡ ln ⁡ t {\displaystyle \delta _{t}:=\ln t+3\ln \ln t} we have the following upper bound on the regret of KL-UCB: R T ≤ ( ∑ a : μ a < μ ∗ Δ a K inf ( ν a , μ ∗ , D ) ) ln ⁡ T + O T ( ln ⁡ T ) . {\displaystyle R_{T}\leq \left(\sum _{a:\mu _{a}<\mu ^{}}{\frac {\Delta _{a}}{{\mathcal {K}}_{\inf }(\nu _{a},\mu ^{},{\mathcal {D}})}}\right)\ln T+O_{T}({\sqrt {\ln T}}).} ==== Bounded distributions in [0,1] ==== For D = P ( [ 0 , 1 ] ) {\displaystyle {\mathcal {D}}={\mathcal {P}}([0,1])} (the set of distributions supported on [ 0 , 1 ] {\displaystyle [0,1]} ), and for δ t := ln ⁡ t + ln ⁡ ln ⁡ t {\displaystyle \delta _{t}:=\ln t+\ln \ln t} , we have the following upper bound on the regret of KL-UCB: R T ≤ ( ∑ a : μ a < μ ∗ Δ a K inf ( ν a , μ ∗ , D ) ) ln ⁡ T + O T ( ( ln ⁡ T ) 4 / 5 ln ⁡ ln ⁡ T ) . {\displaystyle R_{T}\leq \left(\sum _{a:\mu _{a}<\mu ^{}}{\frac {\Delta _{a}}{{\mathcal {K}}_{\inf }(\nu _{a},\mu ^{},{\mathcal {D}})}}\right)\ln T+O_{T}{\big (}(\ln T)^{4/5}\ln \ln T{\big )}.} === Runtime === For D = P ( [ 0 , 1 ] ) {\displaystyle {\mathcal {D}}={\mathcal {P}}([0,1])} , the runtime needed per step and for an arm k {\displaystyle k} with n {\displaystyle n} observations is O ( n ( ln ⁡ n ) 2 ) {\displaystyle {\mathcal {O}}{\big (}n(\ln n)^{2}{\big )}} . This is higher than that of other optimal algorithms, such as NPTS with O ( n ) {\displaystyle {\mathcal {O}}(n)} . MED with O ( n ln ⁡ n ) {\displaystyle {\mathcal {O}}(n\ln n)} . and IMED with O ( n ln ⁡ n ) {\displaystyle {\mathcal {O}}(n\ln n)} . The high runtime of KL-UCB is due to a two-level optimisation: for each arm and candidate mean μ {\displaystyle \mu } , the algorithm evaluates K inf ( ν ^ a ( t ) , μ , D ) {\displaystyle {\mathcal {K}}_{\inf }({\hat {\nu }}_{a}(t),\mu ,{\mathcal {D}})} and then maximises μ {\displaystyle \mu } subject to N a ( t ) K inf ( ν ^ a ( t ) , μ , D ) ≤ δ t {\displaystyle N_{a}(t)\,{\mathcal {K}}_{\inf }({\hat {\nu }}_{a}(t),\mu ,{\mathcal {D}})\leq \delta _{t}} . For distributions bounded in [ 0 , 1 ] {\displaystyle [0,1]} the inner problem has no closed form and must be solved numerically, which increases the per-step cost.

  Read more →