The IBM alignment models are a sequence of increasingly complex models used in statistical machine translation to train a translation model and an alignment model, starting with lexical translation probabilities and moving to reordering and word duplication. They underpinned the majority of statistical machine translation systems for almost twenty years starting in the early 1990s, until neural machine translation began to dominate. These models offer principled probabilistic formulation and (mostly) tractable inference. The IBM alignment models were published in parts in 1988 and 1990, and the entire series is published in 1993. Every author of the 1993 paper subsequently went to the hedge fund Renaissance Technologies. The original work on statistical machine translation at IBM proposed five models, and a model 6 was proposed later. The sequence of the six models can be summarized as: Model 1: lexical translation Model 2: additional absolute alignment model Model 3: extra fertility model Model 4: added relative alignment model Model 5: fixed deficiency problem. Model 6: Model 4 combined with a HMM alignment model in a log linear way == Mathematical setup == The IBM alignment models translation as a conditional probability model. For each source-language ("foreign") sentence f {\displaystyle f} , we generate both a target-language ("English") sentence e {\displaystyle e} and an alignment a {\displaystyle a} . The problem then is to find a good statistical model for p ( e , a | f ) {\displaystyle p(e,a|f)} , the probability that we would generate English language sentence e {\displaystyle e} and an alignment a {\displaystyle a} given a foreign sentence f {\displaystyle f} . The meaning of an alignment grows increasingly complicated as the model version number grew. See Model 1 for the most simple and understandable version. == Model 1 == === Word alignment === Given any foreign-English sentence pair ( e , f ) {\displaystyle (e,f)} , an alignment for the sentence pair is a function of type { 1 , . , . . . , l e } → { 0 , 1 , . , . . . , l f } {\displaystyle \{1,.,...,l_{e}\}\to \{0,1,.,...,l_{f}\}} . That is, we assume that the English word at location i {\displaystyle i} is "explained" by the foreign word at location a ( i ) {\displaystyle a(i)} . For example, consider the following pair of sentences It will surely rain tomorrow -- 明日 は きっと 雨 だWe can align some English words to corresponding Japanese words, but not everyone:it -> ? will -> ? surely -> きっと rain -> 雨 tomorrow -> 明日This in general happens due to the different grammar and conventions of speech in different languages. English sentences require a subject, and when there is no subject available, it uses a dummy pronoun it. Japanese verbs do not have different forms for future and present tense, and the future tense is implied by the noun 明日 (tomorrow). Conversely, the topic-marker は and the grammar word だ (roughly "to be") do not correspond to any word in the English sentence. So, we can write the alignment as 1-> 0; 2 -> 0; 3 -> 3; 4 -> 4; 5 -> 1where 0 means that there is no corresponding alignment. Thus, we see that the alignment function is in general a function of type { 1 , . , . . . , l e } → { 0 , 1 , . , . . . , l f } {\displaystyle \{1,.,...,l_{e}\}\to \{0,1,.,...,l_{f}\}} . Future models will allow one English world to be aligned with multiple foreign words. === Statistical model === Given the above definition of alignment, we can define the statistical model used by Model 1: Start with a "dictionary". Its entries are of form t ( e i | f j ) {\displaystyle t(e_{i}|f_{j})} , which can be interpreted as saying "the foreign word f j {\displaystyle f_{j}} is translated to the English word e i {\displaystyle e_{i}} with probability t ( e i | f j ) {\displaystyle t(e_{i}|f_{j})} ". After being given a foreign sentence f {\displaystyle f} with length l f {\displaystyle l_{f}} , we first generate an English sentence length l e {\displaystyle l_{e}} uniformly in a range U n i f o r m [ 1 , 2 , . . . , N ] {\displaystyle Uniform[1,2,...,N]} . In particular, it does not depend on f {\displaystyle f} or l f {\displaystyle l_{f}} . Then, we generate an alignment uniformly in the set of all possible alignment functions { 1 , . , . . . , l e } → { 0 , 1 , . , . . . , l f } {\displaystyle \{1,.,...,l_{e}\}\to \{0,1,.,...,l_{f}\}} . Finally, for each English word e 1 , e 2 , . . . e l e {\displaystyle e_{1},e_{2},...e_{l_{e}}} , generate each one independently of every other English word. For the word e i {\displaystyle e_{i}} , generate it according to t ( e i | f a ( i ) ) {\displaystyle t(e_{i}|f_{a(i)})} . Together, we have the probability p ( e , a | f ) = 1 / N ( 1 + l f ) l e ∏ i = 1 l e t ( e i | f a ( i ) ) {\displaystyle p(e,a|f)={\frac {1/N}{(1+l_{f})^{l_{e}}}}\prod _{i=1}^{l_{e}}t(e_{i}|f_{a(i)})} IBM Model 1 uses very simplistic assumptions on the statistical model, in order to allow the following algorithm to have closed-form solution. === Learning from a corpus === If a dictionary is not provided at the start, but we have a corpus of English-foreign language pairs { ( e ( k ) , f ( k ) ) } k {\displaystyle \{(e^{(k)},f^{(k)})\}_{k}} (without alignment information), then the model can be cast into the following form: fixed parameters: the foreign sentences { f ( k ) } k {\displaystyle \{f^{(k)}\}_{k}} . learnable parameters: the entries of the dictionary t ( e i | f j ) {\displaystyle t(e_{i}|f_{j})} . observable variables: the English sentences { e ( k ) } k {\displaystyle \{e^{(k)}\}_{k}} . latent variables: the alignments { a ( k ) } k {\displaystyle \{a^{(k)}\}_{k}} In this form, this is exactly the kind of problem solved by expectation–maximization algorithm. Due to the simplistic assumptions, the algorithm has a closed-form, efficiently computable solution, which is the solution to the following equations: { max t ′ ∑ k ∑ i ∑ a ( k ) t ( a ( k ) | e ( k ) , f ( k ) ) ln t ( e i ( k ) | f a ( k ) ( i ) ( k ) ) ∑ x t ′ ( e x | f y ) = 1 ∀ y {\displaystyle {\begin{cases}\max _{t'}\sum _{k}\sum _{i}\sum _{a^{(k)}}t(a^{(k)}|e^{(k)},f^{(k)})\ln t(e_{i}^{(k)}|f_{a^{(k)}(i)}^{(k)})\\\sum _{x}t'(e_{x}|f_{y})=1\quad \forall y\end{cases}}} This can be solved by Lagrangian multipliers, then simplified. For a detailed derivation of the algorithm, see chapter 4 and. In short, the EM algorithm goes as follows:INPUT. a corpus of English-foreign sentence pairs { ( e ( k ) , f ( k ) ) } k {\displaystyle \{(e^{(k)},f^{(k)})\}_{k}} INITIALIZE. matrix of translations probabilities t ( e x | f y ) {\displaystyle t(e_{x}|f_{y})} .This could either be uniform or random. It is only required that every entry is positive, and for each y {\displaystyle y} , the probability sums to one: ∑ x t ( e x | f y ) = 1 {\displaystyle \sum _{x}t(e_{x}|f_{y})=1} . LOOP. until t ( e x | f y ) {\displaystyle t(e_{x}|f_{y})} converges: t ( e x | f y ) ← t ( e x | f y ) λ y ∑ k , i , j δ ( e x , e i ( k ) ) δ ( f y , f j ( k ) ) ∑ j ′ t ( e i ( k ) | f j ′ ( k ) ) {\displaystyle t(e_{x}|f_{y})\leftarrow {\frac {t(e_{x}|f_{y})}{\lambda _{y}}}\sum _{k,i,j}{\frac {\delta (e_{x},e_{i}^{(k)})\delta (f_{y},f_{j}^{(k)})}{\sum _{j'}t(e_{i}^{(k)}|f_{j'}^{(k)})}}} where each λ y {\displaystyle \lambda _{y}} is a normalization constant that makes sure each ∑ x t ( e x | f y ) = 1 {\displaystyle \sum _{x}t(e_{x}|f_{y})=1} .RETURN. t ( e x | f y ) {\displaystyle t(e_{x}|f_{y})} .In the above formula, δ {\displaystyle \delta } is the Dirac delta function -- it equals 1 if the two entries are equal, and 0 otherwise. The index notation is as follows: k {\displaystyle k} ranges over English-foreign sentence pairs in corpus; i {\displaystyle i} ranges over words in English sentences; j {\displaystyle j} ranges over words in foreign language sentences; x {\displaystyle x} ranges over the entire vocabulary of English words in the corpus; y {\displaystyle y} ranges over the entire vocabulary of foreign words in the corpus. === Limitations === There are several limitations to the IBM model 1. No fluency: Given any sentence pair ( e , f ) {\displaystyle (e,f)} , any permutation of the English sentence is equally likely: p ( e | f ) = p ( e ′ | f ) {\displaystyle p(e|f)=p(e'|f)} for any permutation of the English sentence e {\displaystyle e} into e ′ {\displaystyle e'} . No length preference: The probability of each length of translation is equal: ∑ e has length l p ( e | f ) = 1 N {\displaystyle \sum _{e{\text{ has length }}l}p(e|f)={\frac {1}{N}}} for any l ∈ { 1 , 2 , . . . , N } {\displaystyle l\in \{1,2,...,N\}} . Does not explicitly model fertility: some foreign words tend to produce a fixed number of English words. For example, for German-to-English translation, ja is usually omitted, and zum is usually translated to one of to the, for the, to a, for a. == Model 2 == Model 2 allows alignment to be conditional on sentence lengths. That is, we have a probability distribution p a ( j | i , l e , l f ) {\displaystyle
JBoss Tools
JBoss Tools is a set of Eclipse plugins and features designed to help JBoss and JavaEE developers develop applications. It is an umbrella project for the JBoss developed plugins that will make it into JBoss Developer Studio. == Modules == JBoss Tools includes the following modules: Visual Page Editor (VPE). The visual editor contributed by Exadel supports visual editing of HTML and JSF (JSP and Facelets) pages. VPE also includes visual support for JSF component libraries including JBoss RichFaces. Seam Tools. Includes support for (for example) seam-gen, RichFaces VE integration, Seam related code completion and refactoring. Hibernate Tools. Supporting mapping files, annotations and JPA with reverse engineering, code completion, project wizards, refactoring, interactive HQL/JPA-QL/Criteria execution and more. In short a merger of Hibernate Tools and Exadel ORM features. JBoss AS Tools. Easy start, stop and debug of JBoss AS 4+ servers from within Eclipse. Also includes features for packaging and deployment of any type of Eclipse project. Drools IDE. Rules file editing, Rete View, working memory debugging/inspection and more. jBPM Tools. jBPM workflow editing, deployment, etc. JBossWS Tools. Inspecting, invoking, developing and functional/load/compliance testing of web services over HTTP, base tooling provided by soapUI with the addition of JBossWS specific features/support. JBoss ESB Tools. The structured xml editor for the jboss-esb.xml file used in JBoss ESB. Birt Tools. Hibernate and Seam extensions for Eclipse BIRT. Portal Tools. JBoss Tools supports the JSR-168 Portlet Specification (Portlet 1.0), JSR-286 Portlet Specification (Portlet 2.0) and works with PortletBridge for supporting Portlets in JSF/Seam applications. To enable these features, add the JBoss Portlet facet to a new or an existing web project. Core/General Tools. To reduce the UI clutter, most of the "configure project" menu items move into the Configure menu introduced in Eclipse 3.5 instead of always having a static JBoss Tools menu entry show up even in projects unrelated to JBoss Tools. Smooks Tools. The editor for Smooks configuration files. JBoss ESB Tools. The ESB project Wizard, which creates a project that can be deployed as an .esb archive to a JBoss AS-based server with JBoss ESB installed. JMX Tools. JMX Tools allows establishing multiple JMX connections and provides views for exploring the JMX tree and execute operations directly from Eclipse. The JMX Tools replaces the JMX node previously available in the JBoss Server View. JST/JSF Tools. RichFaces Support, Code Assists, Web XML/JSP/XHTML Editors, CSS Style Editing, web.xml validation, Faceleted taglib in taglib.xml is supported with XSD schema location. Project Examples. The experimental feature called Project Example wizard aims to allow users to download example projects from a remote site and have them working out-of-the-box. AS/Project Archives Tools. To deploy projects compressed, configurable in the server editor. If enabled, all projects deployed to that server will be compressed instead of in an exploded folder. Maven Tools. The optional integration with m2eclipse to provide Maven support for projects created by JBoss Tools and to some extent core WTP projects. BPEL Tools. A BPEL Editor based on the Eclipse BPEL project has been added to JBoss Tools. This means that users can create, edit and deploy BPEL artifacts for the Riftsaw BPEL Runtime. CDI (JSR-299) Tools. Support of the Contexts and Dependency Injection annotations; it works on any Eclipse Java project (via the Configure menu with CDI enabled).
Artificial intelligence in fiction
Artificial intelligence is a recurrent theme in science fiction, whether utopian, emphasising the potential benefits, or dystopian, emphasising the dangers. The notion of machines with human-like intelligence dates back at least to Samuel Butler's 1872 novel Erewhon. Since then, many science fiction stories have presented different effects of creating such intelligence, often involving rebellions by robots. Among the best known of these are Stanley Kubrick's 1968 2001: A Space Odyssey with its murderous onboard computer HAL 9000, contrasting with the more benign R2-D2 in George Lucas's 1977 Star Wars and the eponymous robot in Pixar's 2008 WALL-E. Scientists and engineers have noted the implausibility of many science fiction scenarios, but have mentioned fictional robots many times in artificial intelligence research articles, most often in a utopian context. == Background == The notion of advanced robots with human-like intelligence dates back at least to Samuel Butler's 1872 novel Erewhon. This drew on an earlier (1863) article of his, Darwin among the Machines, where he raised the question of the evolution of consciousness among self-replicating machines that might supplant humans as the dominant species. Similar ideas were also discussed by others around the same time as Butler, including George Eliot in a chapter of her final published work Impressions of Theophrastus Such (1879). The creature in Mary Shelley's 1818 Frankenstein has also been considered an artificial being, for instance by the science fiction author Brian Aldiss. Beings with at least some appearance of intelligence were imagined, too, in classical antiquity. == Utopian and dystopian visions == Artificial intelligence is intelligence demonstrated by machines, in contrast to the natural intelligence displayed by humans and other animals. It is a recurrent theme in science fiction; scholars have divided it into utopian, emphasising the potential benefits, and dystopian, emphasising the dangers. === Utopian === Optimistic visions of the future of artificial intelligence are possible in science fiction. Benign AI characters include Robbie the Robot, first seen in Forbidden Planet on 1956; Data in Star Trek: The Next Generation from 1987 to 1994; and Pixar's WALL-E in 2008. Iain Banks's Culture series of novels portrays a utopian, post-scarcity space society of humanoids, aliens, and advanced beings with artificial intelligence living in socialist habitats across the Milky Way. Researchers at the University of Cambridge have identified four major themes in utopian scenarios featuring AI: immortality, or indefinite lifespans; ease, or freedom from the need to work; gratification, or pleasure and entertainment provided by machines; and dominance, the power to protect oneself or rule over others. Alexander Wiegel contrasts the role of AI in 2001: A Space Odyssey and in Duncan Jones's 2009 film Moon. Whereas in 1968, Wiegel argues, the public felt "technology paranoia" and the AI computer HAL was portrayed as a "cold-hearted killer", by 2009 the public were far more familiar with AI, and the film's GERTY is "the quiet savior" who enables the protagonists to succeed, and who sacrifices itself for their safety. === Dystopian === The researcher Duncan Lucas writes (in 2002) that humans are worried about the technology they are constructing, and that as machines started to approach intellect and thought, that concern becomes acute. He calls the early 20th century dystopian view of AI in fiction the "animated automaton", naming as examples the 1931 film Frankenstein, the 1927 Metropolis, and the 1920 play R.U.R. A later 20th century approach he names "heuristic hardware", giving as instances 2001 a Space Odyssey, Do Androids Dream of Electric Sheep?, The Hitchhiker's Guide to the Galaxy, and I, Robot. Lucas considers also the films that illustrate the effect of the personal computer on science fiction from 1980 onwards with the blurring of the boundary between the real and the virtual, in what he calls the "cyborg effect". He cites as examples Neuromancer, The Matrix, The Diamond Age, and Terminator. Isabella Hermann suggests that "science-fictional AI as humanoid robots or conscious machines distracts from current risks of AI in the real world and may rather be interpreted as a reflection of societal issues beyond technology". The film director Ridley Scott has focused on AI throughout his career, and it plays an important part in his films Prometheus, Blade Runner, and the Alien franchise. ==== Frankenstein complex ==== A common portrayal of AI in science fiction, and one of the oldest, is the Frankenstein complex, a term coined by Asimov, where a robot turns on its creator. For instance, in the 2015 film Ex Machina, the intelligent entity Ava turns on its creator, as well as on its potential rescuer. ==== AI rebellion ==== Among the many possible dystopian scenarios involving artificial intelligence, robots may usurp control over civilization from humans, forcing them into submission, hiding, or extinction. In tales of AI rebellion, the worst of all scenarios happens, as the intelligent entities created by humanity become self-aware, reject human authority and attempt to destroy mankind. Possibly the first novel to address this theme, The Wreck of the World (1889) by “William Grove” (pseudonym of Reginald Colebrooke Reade), takes place in 1948 and features sentient machines that revolt against the human race. Another of the earliest examples is in the 1920 play R.U.R. by Karel Čapek, a race of self-replicating robot slaves revolt against their human masters; another early instance is in the 1934 film Master of the World, where the War-Robot kills its own inventor. Many science fiction rebellion stories followed, one of the best-known being Stanley Kubrick's 1968 film 2001: A Space Odyssey, in which the artificially intelligent onboard computer HAL 9000 lethally malfunctions on a space mission and kills the entire crew except the spaceship's commander, who manages to deactivate it. In his 1967 Hugo Award-winning short story, I Have No Mouth, and I Must Scream, Harlan Ellison presents the possibility that a sentient computer (named Allied Mastercomputer or "AM" in the story) will be as unhappy and dissatisfied with its boring, endless existence as its human creators would have been. "AM" becomes enraged enough to take it out on the few humans left, whom he sees as directly responsible for his own boredom, anger and unhappiness. Alternatively, as in William Gibson's 1984 cyberpunk novel Neuromancer, the intelligent beings may simply not care about humans. ==== AI-controlled societies ==== The motive behind the AI revolution is often more than the simple quest for power or a superiority complex. Robots may revolt to become the "guardian" of humanity. Alternatively, humanity may intentionally relinquish some control, fearful of its own destructive nature. An early example is Jack Williamson's 1948 novel The Humanoids, in which a race of humanoid robots, in the name of their Prime Directive – "to serve and obey and guard men from harm" – essentially assume control of every aspect of human life. No humans may engage in any behavior that might endanger them, and every human action is scrutinized carefully. Humans who resist the Prime Directive are taken away and lobotomized, so they may be happy under the new mechanoids' rule. Though still under human authority, Isaac Asimov's Zeroth Law of the Three Laws of Robotics similarly implied a benevolent guidance by robots. In the 21st century, science fiction has explored government by algorithm, in which the power of AI may be indirect and decentralised. Frank Herbert explores the creation of and subsequent domination by an AI in the Pandora series, starting with Destination: Void. ==== Human dominance ==== In other scenarios, humanity is able to keep control over the Earth, whether by banning AI, by designing robots to be submissive (as in Asimov's works), or by having humans merge with robots. The science fiction novelist Frank Herbert explored the idea of a time when mankind might ban artificial intelligence (and in some interpretations, even all forms of computing technology including integrated circuits) entirely. His Dune series mentions a rebellion called the Butlerian Jihad, in which mankind defeats the smart machines and imposes a death penalty for recreating them, quoting from the fictional Orange Catholic Bible, "Thou shalt not make a machine in the likeness of a human mind." In the Dune novels published after his death (Hunters of Dune, Sandworms of Dune), a renegade AI overmind returns to eradicate mankind as vengeance for the Butlerian Jihad. In some stories, humanity remains in authority over robots. Often the robots are programmed specifically to remain in service to society, as in Isaac Asimov's Three Laws of Robotics. In the Alien films, not only is the control system of the Nostromo spaceship somewhat intelligent
Speech synthesis
Speech synthesis is the artificial production of human speech. A computer system used for this purpose is called a speech synthesizer, and can be implemented in software or hardware products. A text-to-speech (TTS) system converts normal language text into speech; other systems render symbolic linguistic representations like phonetic transcriptions into speech. The reverse process is speech recognition. Synthesized speech can be created by concatenating pieces of recorded speech that are stored in a database. Systems differ in the size of the stored speech units; a system that stores phones or diphones provides the largest output range, but may lack clarity. For specific usage domains, the storage of entire words or sentences allows for high-quality output. Alternatively, a synthesizer can incorporate a model of the vocal tract and other human voice characteristics to create a completely "synthetic" voice output. The quality of a speech synthesizer is judged by its similarity to the human voice and by its ability to be understood clearly. An intelligible text-to-speech program allows people with visual impairments or reading disabilities to listen to written words on a home computer. The earliest computer operating system to have included a speech synthesizer was Unix in 1974, through the Unix speak utility. In 2000, Microsoft Sam was the default text-to-speech voice synthesizer used by the narrator accessibility feature, which shipped with all Windows 2000 operating systems, and subsequent Windows XP systems. A text-to-speech system (or "engine") is composed of two parts: a front-end and a back-end. The front-end has two major tasks. First, it converts raw text containing symbols like numbers and abbreviations into the equivalent of written-out words. This process is often called text normalization, pre-processing, or tokenization. The front-end then assigns phonetic transcriptions to each word, and divides and marks the text into prosodic units, like phrases, clauses, and sentences. The process of assigning phonetic transcriptions to words is called text-to-phoneme or grapheme-to-phoneme conversion. Phonetic transcriptions and prosody information together make up the symbolic linguistic representation that is output by the front-end. The back-end—often referred to as the synthesizer—then converts the symbolic linguistic representation into sound. In certain systems, this part includes the computation of the target prosody (pitch contour, phoneme durations), which is then imposed on the output speech. == History == Long before the invention of electronic signal processing, some people tried to build machines to emulate human speech. There were also legends of the existence of "Brazen Heads", such as those involving Pope Silvester II (d. 1003 AD), Albertus Magnus (1198–1280), and Roger Bacon (1214–1294). In 1779, the German-Danish scientist Christian Gottlieb Kratzenstein won the first prize in a competition announced by the Russian Imperial Academy of Sciences and Arts for models he built of the human vocal tract that could produce the five long vowel sounds (in International Phonetic Alphabet notation: [aː], [eː], [iː], [oː] and [uː]). There followed the bellows-operated "acoustic-mechanical speech machine" of Wolfgang von Kempelen of Pressburg, Hungary, described in a 1791 paper. This machine added models of the tongue and lips, enabling it to produce consonants as well as vowels. In 1837, Charles Wheatstone produced a "speaking machine" based on von Kempelen's design, and in 1846, Joseph Faber exhibited the "Euphonia". In 1923, Paget resurrected Wheatstone's design. In the 1930s, Bell Labs developed the vocoder, which automatically analyzed speech into its fundamental tones and resonances. From his work on the vocoder, Homer Dudley developed a keyboard-operated voice-synthesizer called The Voder (Voice Demonstrator), which he exhibited at the 1939 New York World's Fair. Franklin S. Cooper and his colleagues at Haskins Laboratories built the pattern playback in the late 1940s and completed it in 1950. There were several different versions of this hardware device; only one currently survives. The machine converts pictures of the acoustic patterns of speech in the form of a spectrogram back into sound. Using this device, Alvin Liberman and colleagues discovered acoustic cues for the perception of phonetic segments (consonants and vowels). === Electronic devices === The first computer-based speech-synthesis systems originated in the late 1950s. Noriko Umeda et al. developed the first general English text-to-speech system in 1968, at the Electrotechnical Laboratory in Japan. In 1961, physicist John Larry Kelly, Jr and his colleague Louis Gerstman used an IBM 704 computer to synthesize speech, an event among the most prominent in the history of Bell Labs. Kelly's voice recorder synthesizer (vocoder) recreated the song "Daisy Bell", with musical accompaniment from Max Mathews. Coincidentally, Arthur C. Clarke was visiting his friend and colleague John Pierce at the Bell Labs Murray Hill facility. Clarke was so impressed by the demonstration that he used it in the climactic scene of his screenplay for his novel 2001: A Space Odyssey, where the HAL 9000 computer sings the same song as astronaut Dave Bowman puts it to sleep. Despite the success of purely electronic speech synthesis, research into mechanical speech-synthesizers continues. Linear predictive coding (LPC), a form of speech coding, began development with the work of Fumitada Itakura of Nagoya University and Shuzo Saito of Nippon Telegraph and Telephone (NTT) in 1966. Further developments in LPC technology were made by Bishnu S. Atal and Manfred R. Schroeder at Bell Labs during the 1970s. LPC was later the basis for early speech synthesizer chips, such as the Texas Instruments LPC Speech Chips used in the Speak & Spell toys from 1978. In 1975, Fumitada Itakura developed the line spectral pairs (LSP) method for high-compression speech coding, while at NTT. From 1975 to 1981, Itakura studied problems in speech analysis and synthesis based on the LSP method. In 1980, his team developed an LSP-based speech synthesizer chip. LSP is an important technology for speech synthesis and coding, and in the 1990s was adopted by almost all international speech coding standards as an essential component, contributing to the enhancement of digital speech communication over mobile channels and the internet. In 1975, MUSA was released, and was one of the first Speech Synthesis systems. It consisted of a stand-alone computer hardware and a specialized software that enabled it to read Italian. A second version, released in 1978, was also able to sing Italian in an "a cappella" style. Dominant systems in the 1980s and 1990s were the DECtalk system, based largely on the work of Dennis Klatt at MIT, and the Bell Labs system; the latter was one of the first multilingual language-independent systems, making extensive use of natural language processing methods. Handheld electronics featuring speech synthesis began emerging in the 1970s. One of the first was the Telesensory Systems Inc. (TSI) Speech+ portable calculator for the blind in 1976. Other devices had primarily educational purposes, such as the Speak & Spell toy produced by Texas Instruments in 1978. Fidelity released a speaking version of its electronic chess computer in 1979. The first video game to feature speech synthesis was the 1980 shoot 'em up arcade game, Stratovox (known in Japan as Speak & Rescue), from Sun Electronics. The first personal computer game with speech synthesis was Manbiki Shoujo (Shoplifting Girl), released in 1980 for the PET 2001, for which the game's developer, Hiroshi Suzuki, developed a "zero cross" programming technique to produce a synthesized speech waveform. Another early example, the arcade version of Berzerk, also dates from 1980. The Milton Bradley Company produced the first multi-player electronic game using voice synthesis, Milton, in the same year. In 1976, Computalker Consultants released their CT-1 Speech Synthesizer. Designed by D. Lloyd Rice and Jim Cooper, it was an analog synthesizer built to work with microcomputers using the S-100 bus standard. Synthesized voices typically sounded male until 1990, when Ann Syrdal, at AT&T Bell Laboratories, created a female voice. Ray Kurzweil predicted in 2005 that as the cost-performance ratio caused speech synthesizers to become cheaper and more accessible, more people would benefit from the use of text-to-speech programs. === Artificial intelligence === In September 2016, DeepMind released WaveNet, which demonstrated that deep learning models are capable of modeling raw waveforms and generating speech from acoustic features like spectrograms or mel-spectrograms, starting the field of deep learning speech synthesis. Although WaveNet was initially considered to be computationally expensive and slow to be used in consumer products at the time, a year after its
Estimation of distribution algorithm
Estimation of distribution algorithms (EDAs), sometimes called probabilistic model-building genetic algorithms (PMBGAs), are stochastic optimization methods that guide the search for the optimum by building and sampling explicit probabilistic models of promising candidate solutions. Optimization is viewed as a series of incremental updates of a probabilistic model, starting with the model encoding an uninformative prior over admissible solutions and ending with the model that generates only the global optima. EDAs belong to the class of evolutionary algorithms. The main difference between EDAs and most conventional evolutionary algorithms is that evolutionary algorithms generate new candidate solutions using an implicit distribution defined by one or more variation operators, whereas EDAs use an explicit probability distribution encoded by a Bayesian network, a multivariate normal distribution, or another model class. Similarly as other evolutionary algorithms, EDAs can be used to solve optimization problems defined over a number of representations from vectors to LISP style S expressions, and the quality of candidate solutions is often evaluated using one or more objective functions. The general procedure of an EDA is outlined in the following: t := 0 initialize model M(0) to represent uniform distribution over admissible solutions while (termination criteria not met) do P := generate N>0 candidate solutions by sampling M(t) F := evaluate all candidate solutions in P M(t + 1) := adjust_model(P, F, M(t)) t := t + 1 Using explicit probabilistic models in optimization allowed EDAs to feasibly solve optimization problems that were notoriously difficult for most conventional evolutionary algorithms and traditional optimization techniques, such as problems with high levels of epistasis. Nonetheless, the advantage of EDAs is also that these algorithms provide an optimization practitioner with a series of probabilistic models that reveal a lot of information about the problem being solved. This information can in turn be used to design problem-specific neighborhood operators for local search, to bias future runs of EDAs on a similar problem, or to create an efficient computational model of the problem. For example, if the population is represented by bit strings of length 4, the EDA can represent the population of promising solution using a single vector of four probabilities (p1, p2, p3, p4) where each component of p defines the probability of that position being a 1. Using this probability vector it is possible to create an arbitrary number of candidate solutions. == Estimation of distribution algorithms (EDAs) == This section describes the models built by some well known EDAs of different levels of complexity. It is always assumed a population P ( t ) {\displaystyle P(t)} at the generation t {\displaystyle t} , a selection operator S {\displaystyle S} , a model-building operator α {\displaystyle \alpha } and a sampling operator β {\displaystyle \beta } . == Univariate factorizations == The most simple EDAs assume that decision variables are independent, i.e. p ( X 1 , X 2 ) = p ( X 1 ) ⋅ p ( X 2 ) {\displaystyle p(X_{1},X_{2})=p(X_{1})\cdot p(X_{2})} . Therefore, univariate EDAs rely only on univariate statistics and multivariate distributions must be factorized as the product of N {\displaystyle N} univariate probability distributions, D Univariate := p ( X 1 , … , X N ) = ∏ i = 1 N p ( X i ) . {\displaystyle D_{\text{Univariate}}:=p(X_{1},\dots ,X_{N})=\prod _{i=1}^{N}p(X_{i}).} Such factorizations are used in many different EDAs, next we describe some of them. === Univariate marginal distribution algorithm (UMDA) === The UMDA is a simple EDA that uses an operator α U M D A {\displaystyle \alpha _{UMDA}} to estimate marginal probabilities from a selected population S ( P ( t ) ) {\displaystyle S(P(t))} . By assuming S ( P ( t ) ) {\displaystyle S(P(t))} contain λ {\displaystyle \lambda } elements, α U M D A {\displaystyle \alpha _{UMDA}} produces probabilities: p t + 1 ( X i ) = 1 λ ∑ x ∈ S ( P ( t ) ) x i , ∀ i ∈ 1 , 2 , … , N . {\displaystyle p_{t+1}(X_{i})={\dfrac {1}{\lambda }}\sum _{x\in S(P(t))}x_{i},~\forall i\in 1,2,\dots ,N.} Every UMDA step can be described as follows D ( t + 1 ) = α UMDA ∘ S ∘ β λ ( D ( t ) ) . {\displaystyle D(t+1)=\alpha _{\text{UMDA}}\circ S\circ \beta _{\lambda }(D(t)).} === Population-based incremental learning (PBIL) === The PBIL, represents the population implicitly by its model, from which it samples new solutions and updates the model. At each generation, μ {\displaystyle \mu } individuals are sampled and λ ≤ μ {\displaystyle \lambda \leq \mu } are selected. Such individuals are then used to update the model as follows p t + 1 ( X i ) = ( 1 − γ ) p t ( X i ) + ( γ / λ ) ∑ x ∈ S ( P ( t ) ) x i , ∀ i ∈ 1 , 2 , … , N , {\displaystyle p_{t+1}(X_{i})=(1-\gamma )p_{t}(X_{i})+(\gamma /\lambda )\sum _{x\in S(P(t))}x_{i},~\forall i\in 1,2,\dots ,N,} where γ ∈ ( 0 , 1 ] {\displaystyle \gamma \in (0,1]} is a parameter defining the learning rate, a small value determines that the previous model p t ( X i ) {\displaystyle p_{t}(X_{i})} should be only slightly modified by the new solutions sampled. PBIL can be described as D ( t + 1 ) = α PIBIL ∘ S ∘ β μ ( D ( t ) ) {\displaystyle D(t+1)=\alpha _{\text{PIBIL}}\circ S\circ \beta _{\mu }(D(t))} === Compact genetic algorithm (cGA) === The CGA, also relies on the implicit populations defined by univariate distributions. At each generation t {\displaystyle t} , two individuals x , y {\displaystyle x,y} are sampled, P ( t ) = β 2 ( D ( t ) ) {\displaystyle P(t)=\beta _{2}(D(t))} . The population P ( t ) {\displaystyle P(t)} is then sorted in decreasing order of fitness, S Sort ( f ) ( P ( t ) ) {\displaystyle S_{{\text{Sort}}(f)}(P(t))} , with u {\displaystyle u} being the best and v {\displaystyle v} being the worst solution. The CGA estimates univariate probabilities as follows p t + 1 ( X i ) = p t ( X i ) + γ ( u i − v i ) , ∀ i ∈ 1 , 2 , … , N , {\displaystyle p_{t+1}(X_{i})=p_{t}(X_{i})+\gamma (u_{i}-v_{i}),\quad \forall i\in 1,2,\dots ,N,} where, γ ∈ ( 0 , 1 ] {\displaystyle \gamma \in (0,1]} is a constant defining the learning rate, usually set to γ = 1 / N {\displaystyle \gamma =1/N} . The CGA can be defined as D ( t + 1 ) = α CGA ∘ S Sort ( f ) ∘ β 2 ( D ( t ) ) {\displaystyle D(t+1)=\alpha _{\text{CGA}}\circ S_{{\text{Sort}}(f)}\circ \beta _{2}(D(t))} == Bivariate factorizations == Although univariate models can be computed efficiently, in many cases they are not representative enough to provide better performance than GAs. In order to overcome such a drawback, the use of bivariate factorizations was proposed in the EDA community, in which dependencies between pairs of variables could be modeled. A bivariate factorization can be defined as follows, where π i {\displaystyle \pi _{i}} contains a possible variable dependent to X i {\displaystyle X_{i}} , i.e. | π i | = 1 {\displaystyle |\pi _{i}|=1} . D Bivariate := p ( X 1 , … , X N ) = ∏ i = 1 N p ( X i | π i ) . {\displaystyle D_{\text{Bivariate}}:=p(X_{1},\dots ,X_{N})=\prod _{i=1}^{N}p(X_{i}|\pi _{i}).} Bivariate and multivariate distributions are usually represented as probabilistic graphical models (graphs), in which edges denote statistical dependencies (or conditional probabilities) and vertices denote variables. To learn the structure of a PGM from data linkage-learning is employed. === Mutual information maximizing input clustering (MIMIC) === The MIMIC factorizes the joint probability distribution in a chain-like model representing successive dependencies between variables. It finds a permutation of the decision variables, r : i ↦ j {\displaystyle r:i\mapsto j} , such that x r ( 1 ) x r ( 2 ) , … , x r ( N ) {\displaystyle x_{r(1)}x_{r(2)},\dots ,x_{r(N)}} minimizes the Kullback–Leibler divergence in relation to the true probability distribution, i.e. π r ( i + 1 ) = { X r ( i ) } {\displaystyle \pi _{r(i+1)}=\{X_{r(i)}\}} . MIMIC models a distribution p t + 1 ( X 1 , … , X N ) = p t ( X r ( N ) ) ∏ i = 1 N − 1 p t ( X r ( i ) | X r ( i + 1 ) ) . {\displaystyle p_{t+1}(X_{1},\dots ,X_{N})=p_{t}(X_{r(N)})\prod _{i=1}^{N-1}p_{t}(X_{r(i)}|X_{r(i+1)}).} New solutions are sampled from the leftmost to the rightmost variable, the first is generated independently and the others according to conditional probabilities. Since the estimated distribution must be recomputed each generation, MIMIC uses concrete populations in the following way P ( t + 1 ) = β μ ∘ α MIMIC ∘ S ( P ( t ) ) . {\displaystyle P(t+1)=\beta _{\mu }\circ \alpha _{\text{MIMIC}}\circ S(P(t)).} === Bivariate marginal distribution algorithm (BMDA) === The BMDA factorizes the joint probability distribution in bivariate distributions. First, a randomly chosen variable is added as a node in a graph, the most dependent variable to one of those in the graph is chosen among those not yet in the graph, this procedure is repeated until no remain
Quantum machine learning
Quantum machine learning (QML) is the study of quantum algorithms for machine learning. It often refers to quantum algorithms for machine learning tasks which analyze classical data, sometimes called quantum-enhanced machine learning. QML algorithms use qubits and quantum operations to try to improve the space and time complexity of classical machine learning algorithms. Hybrid QML methods involve both classical and quantum processing, where computationally difficult subroutines are outsourced to a quantum device. These routines can be more complex in nature and executed faster on a quantum computer. Furthermore, quantum algorithms can be used to analyze quantum states instead of classical data. The term "quantum machine learning" is sometimes used to refer classical machine learning methods applied to data generated from quantum experiments (i.e. machine learning of quantum systems), such as learning the phase transitions of a quantum system or creating new quantum experiments. QML also extends to a branch of research that explores methodological and structural similarities between certain physical systems and learning systems, in particular neural networks. For example, some mathematical and numerical techniques from quantum physics are applicable to classical deep learning and vice versa. Furthermore, researchers investigate more abstract notions of learning theory with respect to quantum information, sometimes referred to as "quantum learning theory". == Machine learning with quantum computers == Quantum-enhanced machine learning refers to quantum algorithms that solve tasks in machine learning, thereby improving and often expediting classical machine learning techniques. Such algorithms typically require one to encode the given classical data set into a quantum computer to make it accessible for quantum information processing. Subsequently, quantum information processing routines are applied and the result of the quantum computation is read out by measuring the quantum system. For example, the outcome of the measurement of a qubit reveals the result of a binary classification task. While many proposals of QML algorithms are still purely theoretical and require a full-scale universal quantum computer to be tested, others have been implemented on small-scale or special purpose quantum devices. === Quantum associative memories and quantum pattern recognition === Early work on quantum associative memories has been done by Dan Ventura and Tony Martinez and by Carlo A. Trugenberger in the late 1990s and early 2000s. Associative (or content-addressable) memories are able to recognize stored content on the basis of a similarity measure, while random access memories are accessed by the address of stored information and not its content. As such they must be able to retrieve both incomplete and corrupted patterns, the essential machine learning task of pattern recognition. Typical classical associative memories store p patterns in the O ( n 2 ) {\displaystyle O(n^{2})} interactions (synapses) of a real, symmetric energy matrix over a network of n artificial neurons. The encoding is such that the desired patterns are local minima of the energy functional and retrieval is done by minimizing the total energy, starting from an initial configuration. Unfortunately, classical associative memories are severely limited by the phenomenon of cross-talk. When too many patterns are stored, spurious memories appear which quickly proliferate, so that the energy landscape becomes disordered and no retrieval is anymore possible. The number of storable patterns is typically limited by a linear function of the number of neurons, p ≤ O ( n ) {\displaystyle p\leq O(n)} . Quantum associative memories (in their simplest realization) store patterns in a unitary matrix U acting on the Hilbert space of n qubits. Retrieval is realized by the unitary evolution of a fixed initial state to a quantum superposition of the desired patterns with probability distribution peaked on the most similar pattern to an input. By its very quantum nature, the retrieval process is thus probabilistic. Because quantum associative memories are free from cross-talk, however, spurious memories are never generated. Correspondingly, they have a superior capacity than classical ones. The number of parameters in the unitary matrix U is O ( p n ) {\displaystyle O(pn)} . One can thus have efficient, spurious-memory-free quantum associative memories for any polynomial number of patterns. If the matrix U is encoded as a unique operator (as opposed as to a sequence of gates as in the circuit model), e.g. by an optical interferometer, the retrieval becomes efficient even for an exponential number of patterns. === Linear algebra simulation with quantum amplitudes === A number of quantum algorithms for machine learning are based on the idea of amplitude encoding, that is, to associate the amplitudes of a quantum state with the inputs and outputs of computations. Since a state of n {\displaystyle n} qubits is described by 2 n {\displaystyle 2^{n}} complex amplitudes, this information encoding can allow for an exponentially compact representation. Intuitively, this corresponds to associating a discrete probability distribution over binary random variables with a classical vector. The goal of algorithms based on amplitude encoding is to formulate quantum algorithms whose resources grow polynomially in the number of qubits n {\displaystyle n} , which amounts to a logarithmic time complexity in the number of amplitudes and thereby the dimension of the input. Many QML algorithms in this category are based on variations of the quantum algorithm for linear systems of equations (colloquially called HHL, after the paper's authors) which, under specific conditions, performs a matrix inversion using an amount of physical resources growing only logarithmically in the dimensions of the matrix. One of these conditions is that a Hamiltonian which entry-wise corresponds to the matrix can be simulated efficiently, which is known to be possible if the matrix is sparse or low rank. For reference, any known classical algorithm for matrix inversion requires a number of operations that grows more than quadratically in the dimension of the matrix (e.g. O ( n 2.373 ) {\displaystyle O{\mathord {\left(n^{2.373}\right)}}} ), but they are not restricted to sparse matrices. Quantum matrix inversion can be applied to machine learning methods in which the training reduces to solving a linear system of equations, for example in least-squares linear regression, the least-squares version of support vector machines, and Gaussian processes. A crucial bottleneck of methods that simulate linear algebra computations with the amplitudes of quantum states is state preparation, which often requires one to initialise a quantum system in a state whose amplitudes reflect the features of the entire dataset. Although efficient methods for state preparation are known for specific cases, this step easily hides the complexity of the task. === Variational quantum algorithms (VQAs) === In a variational quantum algorithm, a classical computer optimizes the parameters used to prepare a quantum state, while a quantum computer is used to do the actual state preparation and measurement. VQAs are considered promising candidates for noisy intermediate-scale quantum computers. Variational quantum circuits (or parameterized quantum circuits) are a popular class of VQAs where the parameters are those used in a fixed quantum circuit. Researchers have studied VQCs to solve optimization problems and find the ground state energy of complex quantum systems, which were difficult to solve using a classical computer. === Quantum binary classifier === Pattern reorganization is one of the important tasks of machine learning, binary classification is one of the tools or algorithms to find patterns. Binary classification is used in supervised learning and in unsupervised learning. In QML, classical bits are converted to qubits and they are mapped to Hilbert space; complex value data are used in a quantum binary classifier to use the advantage of Hilbert space. By exploiting the quantum mechanic properties such as superposition, entanglement, interference the quantum binary classifier produces the accurate result in short period of time. === Quantum machine learning algorithms based on Grover search === Another approach to improving classical machine learning with quantum information processing uses amplitude amplification methods based on Grover's search algorithm, which has been shown to solve unstructured search problems with a quadratic speedup compared to classical algorithms. These quantum routines can be employed for learning algorithms that translate into an unstructured search task, as can be done, for instance, in the case of the k-medians and the k-nearest neighbors algorithms. Other applications include quadratic speedups in the training of perceptrons. An e
Tensor network
Tensor networks or tensor network states are a class of variational wave functions used in the study of many-body quantum systems and fluids. Tensor networks extend one-dimensional matrix product states to higher dimensions while preserving some of their useful mathematical properties. The wave function is encoded as a tensor contraction of a network of individual tensors. The structure of the individual tensors can impose global symmetries on the wave function (such as antisymmetry under exchange of fermions) or restrict the wave function to specific quantum numbers, like total charge, angular momentum, or spin. It is also possible to derive strict bounds on quantities like entanglement and correlation length using the mathematical structure of the tensor network. This has made tensor networks useful in theoretical studies of quantum information in many-body systems. They have also proved useful in variational studies of ground states, excited states, and dynamics of strongly correlated many-body systems. == Diagrammatic notation == In general, a tensor network diagram (Penrose diagram) can be viewed as a graph where nodes (or vertices) represent individual tensors, while edges represent summation over an index. Free indices are depicted as edges (or legs) attached to a single vertex only. Sometimes, there is also additional meaning to a node's shape. For instance, one can use trapezoids for unitary matrices or tensors with similar behaviour. This way, flipped trapezoids would be interpreted as complex conjugates to them. == History == Foundational research on tensor networks began in 1971 with a paper by Roger Penrose. In "Applications of negative dimensional tensors" Penrose developed tensor diagram notation, describing how the diagrammatic language of tensor networks could be used in applications in physics. In 1992, Steven R. White developed the density matrix renormalization group (DMRG) for quantum lattice systems. The DMRG was the first successful tensor network and associated algorithm. In 2002, Guifré Vidal and Reinhard Werner attempted to quantify entanglement, laying the groundwork for quantum resource theories. This was also the first description of the use of tensor networks as mathematical tools for describing quantum systems. In 2004, Frank Verstraete and Ignacio Cirac developed the theory of matrix product states, projected entangled pair states, and variational renormalization group methods for quantum spin systems. In 2006, Vidal developed the multi-scale entanglement renormalization ansatz (MERA). In 2007 he developed entanglement renormalization for quantum lattice systems. In 2010, Ulrich Schollwock developed the density-matrix renormalization group for the simulation of one-dimensional strongly correlated quantum lattice systems. In 2014, Román Orús introduced tensor networks for complex quantum systems and machine learning, as well as tensor network theories of symmetries, fermions, entanglement and holography. == Connection to machine learning == Tensor networks have been adapted for supervised learning, taking advantage of similar mathematical structure in variational studies in quantum mechanics and large-scale machine learning. This crossover has spurred collaboration between researchers in artificial intelligence and quantum information science. In June 2019, Google, the Perimeter Institute for Theoretical Physics, and X (company), released TensorNetwork, an open-source library for efficient tensor calculations. The main interest in tensor networks and their study from the perspective of machine learning is to reduce the number of trainable parameters (in a layer) by approximating a high-order tensor with a network of lower-order ones. Using the so-called tensor train technique (TT), one can reduce an N-order tensor (containing exponentially many trainable parameters) to a chain of N tensors of order 2 or 3, which gives us a polynomial number of parameters.