The IBM 2780 and the IBM 3780 are devices developed by IBM for performing remote job entry (RJE) and other batch functions over telephone lines; they communicate with the mainframe via Binary Synchronous Communications (BSC or Bisync) and replaced older terminals using synchronous transmit-receive (STR). In addition, IBM has developed workstation programs for the 1130, 360/20, 2922, System/360 other than 360/20, System/370 and System/3. == 2780 Data Transmission Terminal == The 2780 Data Transmission Terminal first shipped in 1967. It consists of: A line printer similar to the IBM 1443 that can print up to 240 lines per minute (lpm), or 300 lpm using an extremely restricted character set. A card reader/punch unit, similar to an IBM 1442, that can read up to 400 cards per minute (cpm) and can punch up to 355 cpm. A line buffer that stores data received or to be transmitted over the communications line. A binary synchronous adapter which controls the flow of data over the communications line. The 2780 is capable of local (offline) card to print operation. It comes in four models: Model 1: Can read punched cards and transmit the data to a remote host computer, and can receive and print data sent by the host. Model 2: Same as Model 1 but adds the ability to punch card data received from the host. Model 3: Can only print data received from the host, but not send data to it. Model 4: Can read and punch card data, but has no printing capabilities. The 2780 uses a dedicated communication line at speeds of 1200, 2000, 2400 or 4800 bits per second. It is a half duplex device, although full duplex lines can be used with some increase in throughput. It can communicate in Transcode (a 6-bit code), 8-bit EBCDIC, or 7-bit ASCII. == 2770 Data Communication System == The 2770, announced in 1969, "was said to surpass all other IBM terminals in the variety of available input-output devices." The 2770 was developed by the IBM General Products Division (GPD) in Rochester, MN. It comes standard with a desktop terminal with keyboard. The printer and other devices (any two in any combination) can be attached to the 2772 Multi-Purpose Control unit. Possible devices include: 50 Magnetic Data Inscriber 545 Card Punch Model 3 (non-printing) or Model 4 (printing) 1017 Paper Tape Reader 1018 Paper Tape Punch 1053 Printer Model 1 1255 Magnetic Character Reader Models 1, 2 or 3 2203 Printer Model A1 or A2 2213 Printer Model 1 or 2 2265 Display Station Model 2 2502 Card Reader Model A1 or A2 5496 Data Recorder == 3780 Data Communications Terminal == In May 1972, IBM announced the IBM 3780, an enhanced version of the 2780. The 3780 was developed by IBM's Data Processing Division (DPD). There is one model, with an optional card punch. The 3780 drops Transcode support and incorporates several performance enhancements. It supports compression of blank fields in data using run-length encoding. It provides the ability to interleave data between devices, introduces double buffering, and adds support for the Wait-before-transmit ACKnowledgement (WACK) and Temporary Text Delay (TTD) Binary Synchronous control characters. The integrated punched card unit can read cards at 600 cards per minute. The integrated printer is rated at 300, 350 or 425 lines per minute based on characters set (63, 52 or 39 characters). The 3781 Card Punch is an optional feature. It punches 160 columns per second, or 91 cards per minute if all 80 columns are punched. The IBM 2780 and 3780 were later emulated on various types of equipment, including eventually the personal computer. A notable early emulation was the DN60, by Digital Equipment Corporation in the late 1970s. == 3770 Data Communications System == In 1974 IBM Data Processing Division (DPD) offered a successor to the 3780, called the 3770 Data Communications System, supporting SDLC, BSC, BSC Multi-leaving and SNA, depending on the configuration. The 3770 is a family of desk console style terminals that offers a variety of keyboard and printer combinations as well as I/O equipment attachment and communications features. The terminals come built into a desk and include the following models: 3771 Communication Terminal (optional card reader, optional card punch, wire matrix printer) Models 1 (40 cps printer), 2 (80 cps printer), and 3 (120 cps printer). 3773 Communication Terminal (diskette, wire matrix printer) Models 1 (40 cps printer), 2 (80 cps printer), and 3 (120 cps printer). Each model has a P version which adds some programming features. 3774 Communication Terminal (optional card reader, optional card punch, optional belt printer, wire matrix printer) Models 1 (80 cps printer), and 2 (120 cps printer). Each model has a P version which adds some programming features, a 480-character display and a non-removable diskette. 3775 Communication Terminal (optional card reader, optional card punch, optional diskette, belt printer) Model 1 (120 lpm printer). The model P1 adds some programming features, a 480-character display and a non-removable diskette. 3776 Communication Terminal (optional card reader, optional card punch, optional diskette, belt printer) Models 1 (300 lpm printer) and 2 (400 lpm printer). Models 3 and 4 are similar to models 1 and 2. 3777 Communication Terminal (optional card reader, optional diskette, train printer) Model 1 (up to 1000 lpm printer depending on character set). Model 2 adds an optional card punch, model 3 adds an optional magnetic tape drive and model 4 replaces the train printer with a slower model called the IBM 3262. The model 4 also allows a second, optional, 3262. The following I/O devices can be attached to a 3770 terminal: IBM 2502 Card Reader: Models A1 (up to 150 card per minute), A2 (up to 300 cards per minute) or A3 (up to 400 cards per minute) IBM 3203 Printer Model 3: 1000 LPM using 48 character set IBM 3501 Card Reader: Up to 50 cards per minute desktop unit IBM 3521 Card Punch: Up to 50 cards per minute IBM 3782 Card Attachment unit, which allows the 2502 or 3521 to be attached to any terminal except the 3777 IBM 3784 Line Printer, can be attached to a 3774 as a second printer. Up to 155 LPM with 48 characters set print belt. == Workstation programs == IBM distributes workstation programs with systems software including OS/360 Attached Support Processor (ASP) Houston Automatic Spooling Priority (HASP and HASP II) Operating System/Virtual Storage 1 (OS/VS1) Operating System/Virtual Storage 2 (OS/VS2 MVS) Release 2 through 3.8 MVS versions from MVS/SP Version 1 through z/OS Priority Output Writers, Execution processors and input Readers (POWER) Remote Spooling Communications Subsystem (RSCS) Except for the RJE workstation programs in OS/360, these programs use a variation of BSC known as Multi-leaving. In addition, IBM provides separately ordered workstation programs using BSC. Systems Network Architecture (SNA) and TCP/IP. Workstation programs are available from IBM and third-party vendors to support all of these protocols: 2770/3770 2780/3780 Multileaving Network Job Entry (NJE) OS/360 RJE SNA TCP/IP
PCVC Speech Dataset
The PCVC (Persian Consonant Vowel Combination) Speech Dataset is a Modern Persian speech corpus for speech recognition and also speaker recognition. The dataset contains sound samples of Modern Persian combination of vowel and consonant phonemes from different speakers. Every sound sample contains just one consonant and one vowel So it is somehow labeled in phoneme level. This dataset consists of 23 Persian consonants and 6 vowels. The sound samples are all possible combinations of vowels and consonants (138 samples for each speaker). The sample rate of all speech samples is 48000 which means there are 48000 sound samples in every 1 second. Every sound sample starts with consonant then continues with vowel. In each sample, in average, 0.5 second of each sample is speech and the rest is silence. Each sound sample ends with silence. All of sound samples are denoised with "Adaptive noise reduction" algorithm. Compared to Farsdat speech dataset and Persian speech corpus it is more easy to use because it is prepared in .mat data files. Also it is more based on phoneme based separation and all samples are denoised. == Contents == The corpus is downloadable from its Kaggle web page, and contains the following: .mat data files of sound samples in a 23630000 matrix, in which 23 is number of consonants, 6 is the number of vowels and 30000 is the length of sound sample.
TD-Gammon
TD-Gammon is a computer backgammon program developed in the 1990s by Gerald Tesauro at IBM's Thomas J. Watson Research Center. Its name comes from the fact that it is an artificial neural net trained by a form of temporal-difference learning, specifically TD-Lambda. It explored strategies that humans had not pursued and led to advances in the theory of correct backgammon play. In 1993, TD-Gammon (version 2.1) was trained with 1.5 million games of self-play, and achieved a level of play just slightly below that of the top human backgammon players of the time. In 1998, during a 100-game series, it was defeated by the world champion by a mere margin of 8 points. Its unconventional assessment of some opening strategies had been accepted and adopted by expert players. TD-gammon is commonly cited as an early success of reinforcement learning and neural networks, and was cited in, for example, papers for deep Q-learning and AlphaGo. == Algorithm for play and learning == During play, TD-Gammon examines on each turn all possible legal moves and all their possible responses (lookahead search), feeds each resulting board position into its evaluation function, and chooses the move that leads to the board position that got the highest score. In this respect, TD-Gammon is no different than almost any other computer board-game program. TD-Gammon's innovation was in how it learned its evaluation function. TD-Gammon's learning algorithm consists of updating the weights in its neural net after each turn to reduce the difference between its evaluation of previous turns' board positions and its evaluation of the present turn's board position—hence "temporal-difference learning". The score of any board position is a set of four numbers reflecting the program's estimate of the likelihood of each possible game result: White wins normally, Black wins normally, White wins a gammon, Black wins a gammon. For the final board position of the game, the algorithm compares with the actual result of the game rather than its own evaluation of the board position. The core of TD-gammon is a neural network with 3 layers. The input layer has two types of neurons. One type codes for the board position. They are non-negative integers ranging from 0 to 15, indicating the number of White or Black checkers at each board location. There are 99 input neurons for each, totaling 198 neurons. Another type codes for hand-crafted features previously used in Neurogammon. These features encoded standard concepts used by human experts, such as "advanced anchor," "blockade strength," "home board strength" and the probability of a "blot" (single checker) being hit. The hidden layer contains hidden neurons. Later versions had more of these. The output layer contains 4 neurons, representing the network's estimate of the probability ("equity") that the current board would lead to. The 4 neurons code for: White normal win, White gammon win, Black normal win, Black gammon win. Backgammon win is so rare that Tesauro opted to not represent it. After each turn, the learning algorithm updates each weight in the neural net according to the following rule: w t + 1 − w t = α ( Y t + 1 − Y t ) ∑ k = 1 t λ t − k ∇ w Y k {\displaystyle w_{t+1}-w_{t}=\alpha (Y_{t+1}-Y_{t})\sum _{k=1}^{t}\lambda ^{t-k}\nabla _{w}Y_{k}} where: It was found that picking small λ {\displaystyle \lambda } offered performance roughly equally good, and large λ {\displaystyle \lambda } degraded performance. Because of this, after 1992, TD-Gammon was trained with λ = 0 {\displaystyle \lambda =0} , degenerating into standard TD-learning. This saved compute by a factor of 2. == Development history == Version 1.0 used simple 1-ply search: every next move is scored by the neural net, and the highest-scoring move is selected. Versions 2.0 and 2.1 used 2-ply search: Make a 1-ply analysis to remove unlikely moves ("forward pruning"). Make a 2-play minimax analysis for only the likely moves. Pick the best move, probability-weighted by each of the opponent's 21 possible dice rolls (weighting non-doubles twice as much as doubles). Versions 3.0 and 3.1 used 3-ply search, using 21 2 = 441 {\displaystyle 21^{2}=441} possible dice rolls instead of 21. The last version, 3.1, was trained specifically for an exhibition match against Malcolm Davis at the 1998 AAAI Hall of Champions. It lost at -8 points, mainly due to one blunder, where TD-Gammon opted to double and got gammoned at -32 points. == Experiments and stages of training == Unlike previous neural-net backgammon programs such as Neurogammon (also written by Tesauro), where an expert trained the program by supplying the "correct" evaluation of each position, TD-Gammon was at first programmed "knowledge-free". In early experimentation, using only a raw board encoding with no human-designed features, TD-Gammon reached a level of play comparable to Neurogammon: that of an intermediate-level human backgammon player. Even though TD-Gammon discovered insightful features on its own, Tesauro wondered if its play could be improved by using hand-designed features like Neurogammon's. Indeed, the self-training TD-Gammon with expert-designed features soon surpassed all previous computer backgammon programs. It stopped improving after about 1,500,000 games (self-play) using a three-layered neural network, with 198 input units encoding expert-designed features, 80 hidden units, and one output unit representing predicted probability of winning. == Advances in backgammon theory == TD-Gammon's exclusive training through self-play (rather than imitation learning) enabled it to explore strategies that humans previously had not considered or had ruled out erroneously. Its success with unorthodox strategies had a significant impact on the backgammon community. Late 1991, Bill Robertie, Paul Magriel, and Malcolm Davis, were invited to play against TD-Gammon (version 1.0). A total of 51 games were played, with TD-Gammon losing at -0.25 ppg. Robertie found TD-Gammon to be at the level of a competent advanced player, and better than any previous backgammon program. Robertie subsequently wrote about the use of TD-Gammon for backgammon study. For example, on the opening play, the conventional wisdom was that given a roll of 2-1, 4-1, or 5-1, White should move a single checker from point 6 to point 5. Known as "slotting", this technique trades the risk of a hit for the opportunity to develop an aggressive position. TD-Gammon found that the more conservative play of splitting 24-23 was superior. Tournament players began experimenting with TD-Gammon's move, and found success. Within a few years, slotting had disappeared from tournament play, replaced by splitting, though in 2006 it made a reappearance for 2-1. Backgammon expert Kit Woolsey found that TD-Gammon's positional judgement, especially its weighing of risk against safety, was superior to his own or any human's. TD-Gammon's excellent positional play was undercut by occasional poor endgame play. The endgame requires a more analytical approach, sometimes with extensive lookahead. TD-Gammon's limitation to two-ply lookahead put a ceiling on what it could achieve in this part of the game. TD-Gammon's strengths and weaknesses were the opposite of symbolic artificial intelligence programs and most computer software in general: it was good at matters that require an intuitive "feel" but bad at systematic analysis. It is also poor at doubling strategies. This is likely due to the fact that the neural network is trained without the doubling cube, with the doubling added by feeding the neural network's cubeless equity estimates into theoretically-based heuristic formulae. This was particularly the case in the 1998 exhibition match, where it played 100 games against Malcolm Davis. A single doubling blunder lost the match. TD-gammon was never commercialized or released to the public in some other form, but it inspired commercial backgammon programs based on neural networks, such as JellyFish (1994) and Snowie (1998).
Defeasible logic
Defeasible logic is a non-monotonic logic proposed by Donald Nute to formalize defeasible reasoning. In defeasible logic, there are three different types of propositions: strict rules specify that a fact is always a consequence of another; defeasible rules specify that a fact is typically a consequence of another; undercutting defeaters specify exceptions to defeasible rules. A priority ordering over the defeasible rules and the defeaters can be given. During the process of deduction, the strict rules are always applied, while a defeasible rule can be applied only if no defeater of a higher priority specifies that it should not.
Parallel terraced scan
The parallel terraced scan is a multi-agent based search technique that is basic to cognitive architectures, such as Copycat, Letter-string, the Examiner, Tabletop, and others. It was developed by John Rehling and Douglas Hofstadter at the Center for Research on Concepts and Cognition at Indiana University, Bloomington. The parallel terraced scan builds on the concepts of the workspace, coderack, conceptual memory, and temperature. According to Hofstadter the parallel and random nature of the processing captures aspects of human cognition.
Lossless join decomposition
In database design, a lossless join decomposition is a decomposition of a relation r {\displaystyle r} into relations r 1 , r 2 {\displaystyle r_{1},r_{2}} such that a natural join of the two smaller relations yields back the original relation. This is central in removing redundancy safely from databases while preserving the original data. Lossless join can also be called non-additive. == Definition == A relation r {\displaystyle r} on schema R {\displaystyle R} decomposes losslessly onto schemas R 1 {\displaystyle R_{1}} and R 2 {\displaystyle R_{2}} if π R 1 ( r ) ⋈ π R 2 ( r ) = r {\displaystyle \pi _{R_{1}}(r)\bowtie \pi _{R_{2}}(r)=r} , that is r {\displaystyle r} is the natural join of its projections onto the smaller schemas. A pair ( R 1 , R 2 ) {\displaystyle (R_{1},R_{2})} is a lossless-join decomposition of R {\displaystyle R} or said to have a lossless join with respect to a set of functional dependencies F {\displaystyle F} if any relation r ( R ) {\displaystyle r(R)} that satisfies F {\displaystyle F} decomposes losslessly onto R 1 {\displaystyle R_{1}} and R 2 {\displaystyle R_{2}} . Decompositions into more than two schemas can be defined in the same way. == Criteria == A decomposition R = R 1 ∪ R 2 {\displaystyle R=R_{1}\cup R_{2}} has a lossless join with respect to F {\displaystyle F} if and only if the closure of R 1 ∩ R 2 {\displaystyle R_{1}\cap R_{2}} includes R 1 ∖ R 2 {\displaystyle R_{1}\setminus R_{2}} or R 2 ∖ R 1 {\displaystyle R_{2}\setminus R_{1}} . In other words, one of the following must hold: ( R 1 ∩ R 2 ) → ( R 1 ∖ R 2 ) ∈ F + {\displaystyle (R_{1}\cap R_{2})\to (R_{1}\setminus R_{2})\in F^{+}} ( R 1 ∩ R 2 ) → ( R 2 ∖ R 1 ) ∈ F + {\displaystyle (R_{1}\cap R_{2})\to (R_{2}\setminus R_{1})\in F^{+}} === Criteria for multiple sub-schemas === Multiple sub-schemas R 1 , R 2 , . . . , R n {\displaystyle R_{1},R_{2},...,R_{n}} have a lossless join if there is some way in which we can repeatedly perform lossless joins until all the schemas have been joined into a single schema. Once we have a new sub-schema made from a lossless join, we are not allowed to use any of its isolated sub-schema to join with any of the other schemas. For example, if we can do a lossless join on a pair of schemas R i , R j {\displaystyle R_{i},R_{j}} to form a new schema R i , j {\displaystyle R_{i,j}} , we use this new schema (rather than R i {\displaystyle R_{i}} or R j {\displaystyle R_{j}} ) to form a lossless join with another schema R k {\displaystyle R_{k}} (which may already be joined (e.g., R k , l {\displaystyle R_{k,l}} )). == Example == Let R = { A , B , C , D } {\displaystyle R=\{A,B,C,D\}} be the relation schema, with attributes A, B, C and D. Let F = { A → B C } {\displaystyle F=\{A\rightarrow BC\}} be the set of functional dependencies. Decomposition into R 1 = { A , B , C } {\displaystyle R_{1}=\{A,B,C\}} and R 2 = { A , D } {\displaystyle R_{2}=\{A,D\}} is lossless under F because R 1 ∩ R 2 = A {\displaystyle R_{1}\cap R_{2}=A} and we have a functional dependency A → B C {\displaystyle A\rightarrow BC} . In other words, we have proven that ( R 1 ∩ R 2 → R 1 ∖ R 2 ) ∈ F + {\displaystyle (R_{1}\cap R_{2}\rightarrow R_{1}\setminus R_{2})\in F^{+}} .
Geoffrey Hinton
Geoffrey Everest Hinton (born 6 December 1947) is a British-Canadian computer scientist, cognitive scientist, cognitive psychologist and Nobel Prize laureate known for his work on artificial neural networks, which earned him the title "the Godfather of AI". He is University Professor Emeritus at the University of Toronto. From 2013 to 2023, he divided his time working for Google Brain and the University of Toronto before publicly announcing his departure from Google in May 2023, citing concerns about the many risks of artificial intelligence (AI) technology. In 2017, he co-founded and became the chief scientific advisor of the Vector Institute in Toronto. With David Rumelhart and Ronald J. Williams, Hinton was co-author of a highly cited paper published in 1986 that popularised the backpropagation algorithm for training multi-layer neural networks, although they were not the first to propose the approach. Hinton is viewed as a leading figure in the deep learning community. The image-recognition milestone of the AlexNet designed in collaboration with his students Alex Krizhevsky and Ilya Sutskever for the ImageNet challenge 2012 was a breakthrough in the field of computer vision. Hinton received the 2018 Turing Award, together with Yoshua Bengio and Yann LeCun for their work on deep learning. They are sometimes referred to as the "Godfathers of Deep Learning" and have continued to give public talks together. He was also awarded, along with John Hopfield, the 2024 Nobel Prize in Physics for "foundational discoveries and inventions that enable machine learning with artificial neural networks". In May 2023, Hinton announced his resignation from Google to be able to "freely speak out about the risks of AI". He has voiced concerns about deliberate misuse by malicious actors, technological unemployment, and existential risk from artificial general intelligence. He noted that establishing safety guidelines will require cooperation among those competing in use of AI in order to avoid the worst outcomes. After receiving the Nobel Prize, he called for urgent research into AI safety to figure out how to control AI systems smarter than humans. == Education == Hinton was born on 6 December 1947 in Wimbledon in the United Kingdom and was educated at Clifton College in Bristol. In 1967, he matriculated as an undergraduate student at King's College, Cambridge and, after switching between different fields such as natural sciences, history of art, and philosophy, eventually graduated with a Bachelor of Arts in experimental psychology in 1970. He spent a year apprenticing carpentry before returning to academic studies. From 1972 to 1975, he continued his study at the University of Edinburgh, where he was awarded a PhD in artificial intelligence in 1978 for research supervised by Christopher Longuet-Higgins, who favored the symbolic AI approach over the neural network approach. == Career == After his PhD, Hinton initially worked at the University of Sussex and at the MRC Applied Psychology Unit. After having difficulty getting funding in Britain, he worked in the US at the University of California, San Diego, and Carnegie Mellon University. He was the founding director of the Gatsby Charitable Foundation Computational Neuroscience Unit at University College London. He is currently University Professor Emeritus in the Department of Computer Science at the University of Toronto, where he has been affiliated since 1987. Upon arrival in Canada, Geoffrey Hinton was appointed at the Canadian Institute for Advanced Research (CIFAR) in 1987 as a Fellow in CIFAR's first research program, Artificial Intelligence, Robotics & Society. In 2004, Hinton and collaborators successfully proposed the launch of a new program at CIFAR, "Neural Computation and Adaptive Perception" (NCAP), which today is named "Learning in Machines & Brains". Hinton would go on to lead NCAP for ten years. Among the members of the program are Yoshua Bengio and Yann LeCun, with whom Hinton would go on to win the ACM A.M. Turing Award in 2018. All three Turing winners continue to be members of the CIFAR Learning in Machines & Brains program. Hinton taught a free online course on Neural Networks on the education platform Coursera in 2012. He co-founded DNNresearch Inc. in 2012 with his two graduate students, Alex Krizhevsky and Ilya Sutskever, at the University of Toronto's department of computer science. In March 2013, Google acquired DNNresearch Inc. for $44 million, and Hinton planned to "divide his time between his university research and his work at Google". In May 2023, Hinton publicly announced his resignation from Google. He explained his decision, saying he wanted to "freely speak out about the risks of AI" and added that part of him now regrets his life's work. Notable former PhD students and postdoctoral researchers from his group include Peter Dayan, Sam Roweis, Max Welling, Richard Zemel, Brendan Frey, Radford M. Neal, Yee Whye Teh, Ruslan Salakhutdinov, Ilya Sutskever, Yann LeCun, Alex Graves, Zoubin Ghahramani, and Peter Fitzhugh Brown. == Research == Hinton's research concerns the use of neural networks for machine learning, memory, perception, and symbol processing. He has written or co-written more than 200 peer-reviewed publications. In the 1980s, Hinton was part of the "Parallel Distributed Processing" group at Carnegie Mellon University, which included notable scientists like Terrence Sejnowski, Francis Crick, David Rumelhart, and James McClelland. This group favoured the connectionist approach during the AI winter. Their findings were published in a two-volume set. The connectionist approach adopted by Hinton suggests that capabilities in areas like logic and grammar can be encoded into the parameters of neural networks, and that neural networks can learn them from data. Symbolists on the other side advocated for explicitly programming knowledge and rules into AI systems. In 1985, Hinton co-invented Boltzmann machines with David Ackley and Terry Sejnowski. His other contributions to neural network research include distributed representations, time delay neural network, mixtures of experts, Helmholtz machines and product of experts. An accessible introduction to Geoffrey Hinton's research can be found in his articles in Scientific American in September 1992 and October 1993. In 1995, Hinton and colleagues proposed the wake-sleep algorithm, involving a neural network with separate pathways for recognition and generation, being trained with alternating "wake" and "sleep" phases. In 2007, Hinton coauthored an unsupervised learning paper titled Unsupervised learning of image transformations. In 2008, he developed the visualization method t-SNE with Laurens van der Maaten.While Hinton was a postdoc at UC San Diego, David Rumelhart, Hinton and Ronald J. Williams applied the backpropagation algorithm to multi-layer neural networks. Their experiments showed that such networks can learn useful internal representations of data. In a 2018 interview, Hinton said that "David Rumelhart came up with the basic idea of backpropagation, so it's his invention." Although this work was important in popularising backpropagation, it was not the first to suggest the approach. Reverse-mode automatic differentiation, of which backpropagation is a special case, was proposed by Seppo Linnainmaa in 1970, and Paul Werbos proposed to use it to train neural networks in 1974. In 2017, Hinton co-authored two open-access research papers about capsule neural networks, extending the concept of "capsule" introduced by Hinton in 2011. The architecture aims to better model part-whole relationships within objects in visual data. In 2021, Hinton presented GLOM, a speculative architecture idea also aiming to improve image understanding by modeling part-whole relationships in neural networks. In 2021, Hinton co-authored a widely cited paper proposing a framework for contrastive learning in computer vision. The technique involves pulling together representations of augmented versions of the same image, and pushing apart dissimilar representations. At the 2022 Conference on Neural Information Processing Systems (NeurIPS), Hinton introduced a new learning algorithm for neural networks that he calls the "Forward-Forward" algorithm. The idea is to replace the traditional forward-backwards passes of backpropagation with two forward passes, one with positive (i.e. real) data and the other with negative data that could be generated solely by the network. The Forward-Forward algorithm is well-suited for what Hinton calls "mortal computation", where the knowledge learned is not transferable to other systems and thus dies with the hardware, as can be the case for certain analog computers used for machine learning. == Honours and awards == Hinton is a Fellow of the US Association for the Advancement of Artificial Intelligence (FAAAI) since 1990. He was elected a Fellow of the Royal Society of Canada (FRSC) in 1996, and then a