Oren Etzioni (born 1964) is Professor Emeritus of Computer Science at the University of Washington, and founding CEO of the Allen Institute for Artificial Intelligence (AI2). Etzioni is a co-founder of Vercept, an AI startup, and founder and CEO of TrueMedia.org, a non-profit dedicated to fighting political deepfakes, which launched in April 2024. He is also the Founder and Technical Director of the AI2 Incubator and a venture partner at the Madrona Venture Group. == Early life and education == Etzioni is the son of Israeli-American intellectual Amitai Etzioni. He was the first student to major in computer science at Harvard University, where he earned a bachelor's degree in 1986. He earned a PhD from Carnegie Mellon University in January, 1991, supervised by Tom M. Mitchell. == University of Washington career == Etzioni joined the University of Washington faculty in 1991, immediately after receiving his PhD. He rose through the ranks to become the Washington Research Foundation Entrepreneurship Professor in Computer Science & Engineering. Etzioni's research has been focused on basic problems in the study of intelligence, machine reading, machine learning and web search. Past projects include Internet Softbots—the study of intelligent agents in the context of real-world software testbeds. In 2003, he started the KnowItAll project for acquiring massive amounts of information from the web. In 2005, he founded and became the director of the university's Turing Center. The center investigated problems in data mining, natural language processing, the Semantic Web and other web search topics. Etzioni coined the term machine reading and helped to create the first commercial comparison shopping agent. He has published over 200 technical papers, and his H-index exceeds 100. == Entrepreneurship == As a faculty member Etzioni was also an active entrepreneur, founding multiple companies and pioneering multiple technologies including MetaCrawler (bought by Infospace), Netbot (bought by Excite in 1997 for $35 million), and ClearForest (bought by Reuters). He founded Farecast, a travel metasearch and price prediction site, which was acquired by Microsoft in 2008 for $115 million. Before founding Farecast, he developed a program originally called Hamlet, that used algorithms to identify patterns in airfare data using data-mining techniques. He also co-founded Decide.com, a website to help consumers make buying decisions using previous price history and recommendations from other users. Decide.com was bought by eBay in September, 2013. Etzioni is also a venture partner at the Madrona Venture Group. He is founder and CEO of TrueMedia.org, a non-profit dedicated to fighting political deepfakes, which launched in April 2024. Etzioni is a co-founder of Vercept, an AI startup formed in 2025. == Founding CEO of AI2 == In September 2013 Etzioni was selected as the Founding CEO of the Allen Institute for Artificial Intelligence by philanthropist Paul G. Allen, and in January 2014 he took a leave of absence from the University of Washington to serve in that role. Etzioni's technical contributions continued at AI2; for example, in 2015, he helped to create the Semantic Scholar search engine. Under Etzioni’s leadership, AI2 grew from zero to over two hundred team members including notable researchers and engineers across several domains of AI. By 2021, its AI2 researchers had published near 700 papers in publications such as AAAI, ACL, CVPR, NeurIPS, and ICLR. Twenty-four of these papers had garnered special-recognition awards. AI2 also offered several key resources and tools to the AI community including the AllenNLP library, Semantic Scholar, and the conservation platforms EarthRanger and Skylight. Ed Lazowska, AI2 Board Member, has stated about Etzioni that he "took the collegial, collaborative culture that he absorbed in his 20+ years as a professor in UW's Allen School and mixed it with the singular focus that drives startups to create an elixir that AI2 folks have been drinking over the last eight years. The result is an exceptional organization of scientists, engineers, and entrepreneurs that's pursuing Paul Allen’s vision of ‘AI for the Common Good’ with extraordinary success.” == Popular press == In addition to his scientific publications, Etzioni has written commentary on AI for The New York Times, Wired, Nature, and other publications. After reading the idea in a book about AI by Brad Smith and Harry Shum, Etzioni has attempted to create an oath for AI practitioners. In 2018, he published what he called a "Hippocratic Oath for artificial intelligence practitioners" in TechCrunch. == Awards and recognition == In 1993, Etzioni received a National Young Investigator Award. In 2003, Etzioni was elected as AAAI Fellow. In 2005, Etzioni received an IJCAI Distinguished Paper Award for "A Probabilistic Model of Redundancy in Information Extraction". In 2007, he received the Robert S. Engelmore Memorial Award. In 2012 Etzioni was featured as GeekWire's "Geek of the Week". In 2013 Etzioni was voted "Geek of the Year" through GeekWire. In 2022, Etzioni received the 2012 ACL Test-of-Time Paper Award. In 2022, Etzioni, along with Ana-Maria Popescu and Henry Kautz, received the ACM Intelligent User Interfaces Most Impact Award for their 2003 paper, "Towards a Theory of Natural Language Interfaces to Databases". == Personal life == Etzioni has three children, and has said in interviews that family is his number one priority. He is married to Ivone Etzioni, and was previously married to Dr. Ruth Etzioni, a biostatistician at the Fred Hutchinson Cancer Center. Outside of his professional career, Etzioni has a wide range of personal interests. He has attended the Burning Man festival, which he described as a valuable way to step outside his comfort zone. His first computer was a TRS-80, and he has described his car’s GPS as his favorite gadget, joking that he has “no sense of direction.” == Selected publications == === Scholarly publications === Etzioni, Oren (July 1994). "A Softbot-based Interface to the Internet" (PDF). Communications of the ACM. Retrieved March 29, 2018. Etzioni, Oren (December 2008). "Open Information Extraction from the Web" (PDF). Communications of the ACM. Retrieved March 29, 2018. Zamir, Oren; Etzioni, Oren (1998). "Web document clustering". Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval. ACM. pp. 46–54. doi:10.1145/290941.290956. ISBN 978-1-58113-015-7. S2CID 244069. Zamir, Oren; Etzioni, Oren (May 1999). "Grouper: a dynamic clustering interface to Web search results". Computer Networks. 31 (11–16): 1361–1374. CiteSeerX 10.1.1.31.8216. doi:10.1016/S1389-1286(99)00054-7. S2CID 206134308. Popescu, Ana-Maria; Etzioni, Oren (2005). "Extracting product features and opinions from reviews". Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing - HLT '05. pp. 339–346. doi:10.3115/1220575.1220618. Etzioni, Oren; Cafarella, Michael; Downey, Doug; Popescu, Ana-Maria; Shaked, Tal; Sonderland, Stephen; Weld, Daniel; Yates, Alexander (June 2005). "Unsupervised named-entity extraction from the Web: An experimental study". Artificial Intelligence. 165 (1): 91–134. doi:10.1016/j.artint.2005.03.001. Downey, Doug; Etzioni, Oren; Sonderland, Stephen (July 2010). "Grouper: Analysis of a probabilistic model of redundancy in unsupervised information extraction". Artificial Intelligence. 174 (11): 726–748. CiteSeerX 10.1.1.174.2441. doi:10.1016/j.artint.2010.04.024. === Popular articles === Etzioni, Oren (August 4, 2011). "Web Search Needs a Shakeup" (PDF). Nature. Retrieved November 21, 2019. Etzioni, Oren (December 9, 2014). "AI Won't Exterminate Us – It Will Empower Us". Backchannel. Retrieved March 29, 2018. Etzioni, Oren (February 4, 2016). "To Keep AI Safe -- Use AI". Vox. Retrieved November 21, 2019. Etzioni, Oren (April 8, 2016). "Quora Session with Oren Etzioni". Quora. Retrieved March 29, 2018. Etzioni, Oren (June 15, 2016). "Deep Learning Isn't a Dangerous Magic Genie. It's Just Math". Wired. Retrieved March 29, 2018. Etzioni, Oren (September 20, 2016). "No, the Experts Don't Think Superintelligent AI is a Threat to Humanity". MIT Technology Review. Retrieved November 21, 2019. Etzioni, Oren (July 6, 2017). "Artificial intelligence: AI Zooms in on highly influential citations". Nature. Retrieved March 29, 2018. Etzioni, Oren (September 1, 2017). "How to Regulate Artificial Intelligence". The New York Times. Retrieved March 29, 2018. Etzioni, Oren (November 2, 2017). "Workers Displaced by Automation Should Try A New Job: Caregiver". Wired. Retrieved March 29, 2018. Etzioni, Oren (March 14, 2018). "A Hippocratic Oath for artificial intelligence practitioners". Tech Crunch. Retrieved March 29, 2018. Etzioni, Oren (March 7, 2018). "A 'Manhattan Project' for science research". The Hill. Retrieved November 21, 2019. Etzioni, Ore
Video Super Resolution
RTX Video Super Resolution (RTX VSR) is a video scaling feature by Nvidia. It was released on February 28, 2023. == History == The feature was first unveiled during CES 2023 as RTX Video Super Resolution. It uses the on-board Tensor Cores to upscale browser video content in real time. Video Super Resolution was initially only available on RTX 30 and 40 series GPUs, while support for 20 series GPUs was added afterwards; it is now available on all Nvidia RTX-branded GPUs. The feature supports input resolutions from 360p to 1440p and a max output of 4K and comes without support for HDR content although that could be likely added in the future. Nvidia released RTX Video Super Resolution 1.5 with improved video quality and RTX 20 series support on October 17, 2023. == Reception == According to ComputerBase, although "the algorithm is not yet working flawlessly", the feature is "overall recommendable".
Artificial reproduction
Artificial reproduction is the re-creation of life brought about by means other than natural ones. It is new life built by human plans and projects. Examples include artificial selection, artificial insemination, in vitro fertilization, artificial womb, artificial cloning, and kinematic replication. Artificial reproduction is one aspect of artificial life. Artificial reproduction can be categorized into one of two classes according to its capacity to be self-sufficient: non-assisted reproductive technology and assisted reproductive technology. Cutting plants' stems and placing them in compost is a form of assisted artificial reproduction, xenobots are an example of a more autonomous type of reproduction, while the artificial womb presented in the movie the Matrix illustrates a non assisted hypothetical technology. The idea of artificial reproduction has led to various technologies. == Theology == Humans have aspired to create life since immemorial times. Most theologies and religions have conceived this possibility as exclusive of deities. Christian religions consider the possibility of artificial reproduction, in most cases, as heretical and sinful. == Philosophy == Although ancient Greek philosophy raised the concept that man could imitate the creative capacity of nature, classic Greeks thought that if possible, human beings would reproduce things as nature does, and vice versa, nature would do the things that man does in the same way. Aristotle, for example, wrote that if nature made tables, it would make them just as men do. In other words, Aristotle said that if nature were to create a table, such table will look like a human-made table. Correspondingly, Descartes envisioned the human body, and nature, as a machine. Cartesian philosophy does not stop seeing a perfect mirror between nature and the artificial. However, Kant revolutionized this old idea by criticizing such naturalism. Kant pedagogically wrote: "Reason, in order to be taught by nature, must approach nature with its principles in one hand, according to which the agreement among appearances can count as laws, and, in the other hand, the experiment thought out in accord with these principles—in order to be instructed by nature not like a pupil, who has recited to him whatever the teacher wants to say, but like an appointed judge who compels witnesses to answer the questions he puts to them.". Humans are not instructed by nature but rather use nature as raw material to invent. Humans find alternatives to the natural restrictions imposed by natural laws thus, nature is not necessarily mirrored. In accordance with Kant (and contrary to what Aristotle thought) Karl Marx, Alfred Whitehead, Jaques Derrida and Juan David García Bacca noticed that nature is incapable of reproducing tables; or airplanes, or submarines, or computers. If nature tried to create airplanes, it would produce birds. If nature tried to create submarines, it would get fishes. If nature tried to create computers, brains would grow. And if nature tried to create man, modern man, monkeys will be evolved. According to Whitehead, if we look for something natural in artificial life, in the most elaborate cases, if anything, only atoms remain natural. Juan David Garcia Bacca summarized, “It will not come out from wood, it will not be born, a galley; from clay, a vessel; from linen, a dress; from iron, a lever,...From natural, artificial. In the artificial, the natural is reduced to a simple raw material, even though it is perfectly specified with natural specification. The artificial is the real, positive, and original negation of the natural: of species, of genus and of essence. Thus, its ontology is superior to natural ontology. And for this very reason Marx did not attach any importance to Darwin, whose evolutionism is confined to the natural order: to changes, at most, from variety to variety, from species to species... natural. For the same reason, nature has no dialectics, even though continuous evolution and selection can occur. The dialectic cannot emerge from the natural, for deeper reasons than, using today's terms, from a bird, an airplane cannot emerge; from fish, a submarine; from ears, a telephone; from eyes, a television; from a brain, a digital computer; from feet, a car; from hands, an engine; from Euclid, Descartes; from Aristotle, Newton; from Plato, Marx.” According to García Bacca, the major difference between natural causes and artificial causes is that nature does not have plans and projects, while humans design things following plans and projects. In contrast, other influential authors such as Michael Behe have depicted the concept and promoted the idea of intelligent design, a notion that has aroused several doubts and heated controversies, as it reframe natural causes in accordance with a natural plan. Previous ideas that have also provided a positive 'sense' to natural reproduction, are orthogenesis, syntropy, orgone and morphic resonance, among others. Although, these ideas have been historically marginalized and often called pseudoscience, recently Bio-semioticians are reconsidering some of them under symbolic approaches. Current metaphysics of science actually recognizes that the artificial ways of reproduction are diverse from nature, i.e., unnatural, anti-natural or supernatural. Because Biosemiotics does not focus on the function of life but on its meaning, it has a better understanding of the artificial than classic biology. == Science == Biology, being the study of cellular life, addresses reproduction in terms of growth and cellular division (i.e., binary fission, mitosis and meiosis); however, the science of artificial reproduction is not restricted by the mirroring of these natural processes.The science of artificial reproduction is actually transcending the natural forms, and natural rules, of reproduction. For example, xenobots have redefined the classical conception of reproduction. Although xenobots are made of eukariotic cells they do not reproduce by mitosis, but rather by kinematic replication. Such constructive replication does not involve growing but rather building. == Assisted reproductive technologies == Assisted reproductive technology (ART)'s purpose is to assist the development of a human embryo, commonly because of medical concerns due to fertility limitations. == Non-assisted reproductive technologies == Non-assisted reproductive technologies (NART) could have medical motivations but are mostly driven by a wider heterotopic ambition. Although, NARTs are initially designed by humans, they are programed to become independent of humans to a relative or absolute extent. James Lovelock proposed that such novelties could overcome humans. === Artificial cloning === Cloning is the cellular reproductive processes where two or more genetically identical organisms are created, either by natural or artificial means. Artificial cloning normally involves editing the genetic code, somatic cell nuclear transfer and 3D bioprinting. === Non-assisted artificial womb === A non-assisted artificial womb or artificial uterus is a device that allow for ectogenesis or extracorporeal pregnancy by growing an embryonic form outside the body of an organism (that would normally carry the embryo to term) without any human assistance. The aspect of non-assistance is the key distinction between the current artificial womb technology (AWT) in modern medical research, which still relies on human assistance. With this non-assisted hypothetical technology, a zygote or stem cells are used to create an embryo that is then incubated and monitored by artificial intelligence (AI) within a chamber composed of biocompatible material. The AI maintains the necessary conditions for the embryo to develop and thrive, proceeding to mimic organic labor and childbirth in order to best help the embryo adjust to the outside world. Ectogenesis—gestation, depicted in the science fiction movie The Matrix, is a fast approaching reality. This type of innovation presupposes that vertebrate wombs are not the only way for bearing humans or other similar forms of life. === Kinematic replication === Self-replication without binary fission, meiosis, mitosis (or any other form of cellular reproduction that involves division and growing) can be achieved. Xenobots are an example of kinematic replication. They are biobots, named after the African clawed frog (Xenopus laevis). Xenobots are cellular life forms designed by using artificial intelligence to build more of themselves by combining frog cells in a liquid medium. The term kinematic replication is usually reserved for biomolecules (e.g. DNA, RNA, prions, etc.) and artificially designed cellular forms (e.g. xenobots). === Machine constructive replication === Machine constructive replication mimics human traditional manufacturing but is entirely self-automated. Such constructive replication is a more general form of kinematic replication, which does not necessarily
Artificial intelligence in spirituality
Some users of artificial intelligence (AI) technologies, especially chatbots, may develop beliefs that AI has or can attain supernatural or spiritual powers. AI models such as ChatGPT are turned to for fortune telling, mysticism and remote viewing. Recent and sudden advances in large language models have led to folk myths about their origin or capabilities, as well as their deification or worship by some users. Tucker Carlson has made similar claims, including directly to Sam Altman. Pope Leo XIV advised priests against using LLM models when it came to the creation of sermons.
Structured sparsity regularization
Structured sparsity regularization is a class of methods, and an area of research in statistical learning theory, that extend and generalize sparsity regularization learning methods. Both sparsity and structured sparsity regularization methods seek to exploit the assumption that the output variable Y {\displaystyle Y} (i.e., response, or dependent variable) to be learned can be described by a reduced number of variables in the input space X {\displaystyle X} (i.e., the domain, space of features or explanatory variables). Sparsity regularization methods focus on selecting the input variables that best describe the output. Structured sparsity regularization methods generalize and extend sparsity regularization methods, by allowing for optimal selection over structures like groups or networks of input variables in X {\displaystyle X} . Common motivation for the use of structured sparsity methods are model interpretability, high-dimensional learning (where dimensionality of X {\displaystyle X} may be higher than the number of observations n {\displaystyle n} ), and reduction of computational complexity. Moreover, structured sparsity methods allow to incorporate prior assumptions on the structure of the input variables, such as overlapping groups, non-overlapping groups, and acyclic graphs. Examples of uses of structured sparsity methods include face recognition, magnetic resonance image (MRI) processing, socio-linguistic analysis in natural language processing, and analysis of genetic expression in breast cancer. == Definition and related concepts == === Sparsity regularization === Consider the linear kernel regularized empirical risk minimization problem with a loss function V ( y i , f ( x ) ) {\displaystyle V(y_{i},f(x))} and the ℓ 0 {\displaystyle \ell _{0}} "norm" as the regularization penalty: min w ∈ R d 1 n ∑ i = 1 n V ( y i , ⟨ w , x i ⟩ ) + λ ‖ w ‖ 0 , {\displaystyle \min _{w\in \mathbb {R} ^{d}}{\frac {1}{n}}\sum _{i=1}^{n}V(y_{i},\langle w,x_{i}\rangle )+\lambda \|w\|_{0},} where x , w ∈ R d {\displaystyle x,w\in \mathbb {R^{d}} } , and ‖ w ‖ 0 {\displaystyle \|w\|_{0}} denotes the ℓ 0 {\displaystyle \ell _{0}} "norm", defined as the number of nonzero entries of the vector w {\displaystyle w} . f ( x ) = ⟨ w , x i ⟩ {\displaystyle f(x)=\langle w,x_{i}\rangle } is said to be sparse if ‖ w ‖ 0 = s < d {\displaystyle \|w\|_{0}=s
Luminoso
Luminoso is a Cambridge, MA-based text analytics and artificial intelligence company. It spun out of the MIT Media Lab and its crowd-sourced Open Mind Common Sense (OMCS) project. The company has raised $20.6 million in financing, and its clients include Sony, Autodesk, Scotts Miracle-Gro, and GlaxoSmithKline. == History == Luminoso was co-founded in 2010 by Dennis Clark, Jason Alonso, Robyn Speer, and Catherine Havasi, a research scientist at MIT in artificial intelligence and computational linguistics. The company builds on the knowledge base of MIT’s Open Mind Common Sense (OMCS) project, co-founded in 1999 by Havasi, who continues to serve as its director. The OCMS knowledge base has since been combined with knowledge from other crowdsourced resources to become ConceptNet. ConceptNet consists of approximately 28 million statements in 304 languages, with full support for 10 languages and moderate support for 77 languages. ConceptNet is a resource for making an AI that understands the meanings of the words people use. During the World Cup in June 2014, the company provided a widely reported real-time sentiment analysis of the U.S. vs. Germany match, analyzing 900,000 posts on Twitter, Facebook and Google+. == Applications == The company uses artificial intelligence, natural language processing, and machine learning to derive insights from unstructured data such as contact center interactions, chatbot and live chat transcripts, product reviews, open-ended survey responses, and email. Luminoso's software identifies and quantifies patterns and relationships in text-based data, including domain-specific or creative language. Rather than human-powered keyword searches of data, the software automates taxonomy creation around concepts, allowing related words and phrases to be dynamically generated and tracked. Commercial applications include analyzing, prioritizing, and routing contact center interactions; identifying consumer complaints before they begin to trend; and tracking sentiment during product launches. The software natively analyzes text in fourteen languages, as well as emoji. == Products == Luminoso's technology can be accessed via two products: Luminoso Daylight and Luminoso Compass. Luminoso Daylight enables a deep-dive analysis into batch or real-time data, whereas Luminoso Compass automates the categorization of real-time data. Both products offer a user interface as well as an API. Luminoso's products can be implemented through either a cloud-based or an on-premise solution. == Research == Luminoso continues to actively conduct research in natural language processing and word embeddings and regularly participates in evaluations such as SemEval. At SemEval 2017, Luminoso participated in Task 2, measuring the semantic similarity of word pairs within and across five languages. Its solution outperformed all competing systems in every language pair tested, with the exception of Persian. == Recognition == Luminoso has been listed as a "Cool Vendor in AI for Marketing" by Gartner, and has also been named a "Boston Artificial Intelligence Startup to Watch" by BostInno. In May 2017, Luminoso was recognized as having the Best Application for AI in the Enterprise by AI Business, and was also shortlisted as the Best AI Breakthrough and Best Innovation in NLP. == Competitors == Major competitors include Clarabridge and Lexalytics. == Investors == The company raised $1.5 million from angel investors led by Basis Technology in 2012. Its first institutional funding round of $6.5 was completed in July 2014, led by Acadia Woods with participation from Japan’s Digital Garage. The company followed that with a $10M series B funding round in December 2018, led by DVI Equity Partners, with participation from Liberty Global Ventures, DF Enterprises, Raptor Holdco, Acadia Woods Partners, and Accord Ventures, among others.
ELMo
ELMo (embeddings from language model) is a word embedding method for representing a sequence of words as a corresponding sequence of vectors. It was created by researchers at the Allen Institute for Artificial Intelligence, and University of Washington and first released in February 2018. It is a bidirectional LSTM which takes character-level as inputs and produces word-level embeddings, trained on a corpus of about 30 million sentences and 1 billion words. The architecture of ELMo accomplishes a contextual understanding of tokens. Deep contextualized word representation is useful for many natural language processing tasks, such as coreference resolution and polysemy resolution. ELMo was historically important as a pioneer of self-supervised generative pretraining followed by fine-tuning, where a large model is trained to reproduce a large corpus, then the large model is augmented with additional task-specific weights and fine-tuned on supervised task data. It was an instrumental step in the evolution towards transformer-based language modelling. == Architecture == ELMo is a multilayered bidirectional LSTM on top of a token embedding layer. The output of all LSTMs concatenated together consists of the token embedding. The input text sequence is first mapped by an embedding layer into a sequence of vectors. Then two parts are run in parallel over it. The forward part is a 2-layered LSTM with 4096 units and 512 dimension projections, and a residual connection from the first to second layer. The backward part has the same architecture, but processes the sequence back-to-front. The outputs from all 5 components (embedding layer, two forward LSTM layers, and two backward LSTM layers) are concatenated and multiplied by a linear matrix ("projection matrix") to produce a 512-dimensional representation per input token. ELMo was pretrained on a text corpus of 1 billion words. The forward part is trained by repeatedly predicting the next token, and the backward part is trained by repeatedly predicting the previous token. After the ELMo model is pretrained, its parameters are frozen, except for the projection matrix, which can be fine-tuned to minimize loss on specific language tasks. This is an early example of the pretraining-fine-tune paradigm. The original paper demonstrated this by improving state of the art on six benchmark NLP tasks. === Contextual word representation === The architecture of ELMo accomplishes a contextual understanding of tokens. For example, the first forward LSTM of ELMo would process each input token in the context of all previous tokens, and the first backward LSTM would process each token in the context of all subsequent tokens. The second forward LSTM would then incorporate those to further contextualize each token. Deep contextualized word representation is useful for many natural language processing tasks, such as coreference resolution and polysemy resolution. For example, consider the sentenceShe went to the bank to withdraw money.In order to represent the token "bank", the model must resolve its polysemy in context. The first forward LSTM would process "bank" in the context of "She went to the", which would allow it to represent the word to be a location that the subject is going towards. The first backward LSTM would process "bank" in the context of "to withdraw money", which would allow it to disambiguate the word as referring to a financial institution. The second forward LSTM can then process "bank" using the representation vector provided by the first backward LSTM, thus allowing it to represent it to be a financial institution that the subject is going towards. == Historical context == ELMo is one link in a historical evolution of language modelling. Consider a simple problem of document classification, where we want to assign a label (e.g., "spam", "not spam", "politics", "sports") to a given piece of text. The simplest approach is the "bag of words" approach, where each word in the document is treated independently, and its frequency is used as a feature for classification. This was computationally cheap but ignored the order of words and their context within the sentence. GloVe and Word2Vec built upon this by learning fixed vector representations (embeddings) for words based on their co-occurrence patterns in large text corpora. Like BERT (but unlike "bag of words" such as Word2Vec and GloVe), ELMo word embeddings are context-sensitive, producing different representations for words that share the same spelling. It was trained on a corpus of about 30 million sentences and 1 billion words. Previously, bidirectional LSTM was used for contextualized word representation. ELMo applied the idea to a large scale, achieving state of the art performance. After the 2017 publication of Transformer architecture, the architecture of ELMo was changed from a multilayered bidirectional LSTM to a Transformer encoder, giving rise to BERT. BERT has a similar pretrain-fine-tune workflow, but uses a Transformer with implications for more parallelizable training.