User:Yurisugano/sandbox

(Disclaimer: the draft below reflects the current Wikipedia page, which includes the edits made after midterm peer review, but also has edits contributed by other Wikipedia users. My changes as related to the final are all in bold)

Meta AI is an artificial intelligence laboratory that belongs to Meta Platform Inc. (formerly known as Facebook, Inc.) Meta AI seeks to develop artificial intelligence in the digital world, enhancing its augmented and artificial reality technologies. Meta AI is an academic research laboratory focused on generating knowledge for the broader AI community. This is in contrast to Facebook's Applied Machine Learning (AML) team, which focuses on practical applications on its products.

Facebook Artificial Intelligence Research
Meta AI started as Facebook Artificial Intelligence Research (FAIR) with locations in the Menlo Park, California headquarters, London, United Kingdom and a new laboratory in Manhattan. FAIR was officially announced in September, 2013. FAIR was initially directed by New York University's Yann LeCun, a deep learning Professor and Turing Award winner. Working with NYU's Center for Data Science, FAIR's initial goal was to research data science, machine learning, and artificial intelligence. FAIR's goal was "to understand intelligence, to discover its fundamental principles, and to make machines significantly more intelligent". Research at FAIR pioneered the technology that led to face recognition, tagging in photographs, and personalized feed recommendation. In 2014, Vladimir Vapnik joined FAIR. Vapnik is a pioneer in statistical learning, the co-inventor of the support-vector machine, and one of the developers of the Vapnik–Chervonenkis theory.

FAIR opened a research center in Pairs, France in 2015, and subsequently launched smaller satellite research labs in Seattle, Pittsburgh, Tel Aviv, Montreal and London. In 2016, FAIR partnered with Google, Amazon, IBM, and Microsoft in creating the Partnership on Artificial Intelligence to Benefit People and Society, an organization with a focus on open licensed research, supporting ethical and efficient research practices, and discussing fairness, inclusivity, and transparency.

In 2018, Jérôme Pesenti, former CTO of IBM's big data group, assumed the role of president of FAIR, while LeCun stepped down to serve as chief AI scientist. In 2018, FAIR was placed 25th in the AI Research Rankings 2019, which ranked the top global organizations leading AI research. FAIR quickly rose to eighth position in 2019, and maintained eighth position in the 2020 rank. FAIR had approximately 200 staff in 2018, and had the goal to double that number by 2020.

FAIR's initial work included research in learning-model enabled memory networks, self-supervised learning and generative adversarial networks, text classification and translation, as well as computer vision. FAIR released Torch deep-learning modules and in 2017, FAIR released PyTorch, an open-source machine learning framework. PyTorch was subsequently used in several deep learning technologies, such as Tesla's autopilot and Uber's Pyro. Also in 2017, FAIR discontinued a research project once AI bots developed a language that was unintelligible to humans, inciting conversations about dystopian fear of artificial intelligence going out of control. However, FAIR clarified that the research had been shut down because they had accomplished their initial goal to understand how languages are generated, rather than out of fear.

FAIR was renamed Meta AI following the rebranding that changed Facebook, Inc to Meta Platforms Inc.

Current research
In the February 23, 2022 live event Inside the Lab: Building for the Metaverse with AI, the Meta AI team discussed the major advancements in research and development in artificial intelligence. One such tool is the BuilderBot which allows users to generate virtual worlds by using voice commands. Other tools include the No Language Left Behind, a system capable of automatic translation between written languages, and a Universal Speech Translator, a system capable of instantaneous speech-to-speech translation.

Computer vision
Meta AI's computer vision research aims to extract information about the environment from digital images and videos. One example of computer vision technology developed by AI is panoptic segmentation, which recognizes objects in the foreground but also classifies the scenes in the background. Meta AI seeks to improve Visual Question Answering technology, in which a machine answers human user questions about images using cycle-consistency, having the machine generate a question in addition to the answer to address linguistic variations in the questions.

Natural language processing and conversational AI
Artificial intelligence communication requires a machine to understand natural language and to generate language that is natural. Meta AI seeks to improve these technologies to improve safe communication regardless of what language the user might speak. Thus, a central task involves the generalization of natural language processing (NLP) technology to other languages. As such, Meta AI actively works on unsupervised machine translation. Meta AI seeks to improve natural-language interfaces by developing aspects of chitchat dialogue such as repetition, specificity, response-relatedness and question-asking, incorporating personality into image captioning, and generating creativity-based language.

In 2018, Meta AI launched the open-source PyText, a modeling framework focused on NLP systems.

Ranking and recommendations
Meta AI research in ranking & recommendations is behind Facebook's News Feed, Instagram, and individually-recommended Ads and Search results. '''Ranking and recommendations is based on reasoning systems, a technology that aims to implement real-world decision-making. Reasoning systems rely on reinforcement learning, which in turn rely on performance feedback. These processes are difficult to automate. Meta Ai introduced ReAgent, an open source toolset that includes decision-making models, feedback generators and evaluators, and models to automate deployment of decision-making models to real-world applications. In addition to research and recommendations, these technologies are implemented in AI more broadly. For instance, the technologies in ReAgent have been used to teach robots how to learn to move autonomously and to teach machines how to play Go. ReAgent is an expansion of the Horizon program, the first open source reinforcement learning platform. The Horizon program was the earliest culmination of Meta AI's efforts in producing reasoning systems.'''

Systems research
Machine learning and AI depend on the development of novel algorithms, software and hardware technologies. As such, Meta AI's systems research teams studies computer languages, compiler, and hardware applications. '''The systems research team addresses fundamental issues with computing such as floating-point error mitigation. The team is also responsible for evaluation of novel algorithm performance on the most recently developed technology, such as GPUs .'''

Theory
Meta AI studies the mathematical and theoretical foundation to artificial intelligence. Meta AI has publications in learning theory, optimization, and signal processing.

Everything below is related to the midterm submission
Meta AI is an artificial intelligence laboratory that belongs to Meta Platform Inc. (formerly known as Facebook, Inc. ). Meta AI seeks to develop artificial intelligence in the digital world, enhancing its augmented and artificial reality technologies. Meta AI is an academic research laboratory focused on generating knowledge for the broader AI community, and operates in parallel to Facebook's Applied Machine Learning (AML) team, which focuses on practical applications on its products.

History
MetaAI started as Facebook Artificial Intelligence Research (FAIR) with locations in the Menlo Park, California headquarters, London, United Kingdom and a new laboratory in Manhattan. FAIR was officially announced in September, 2013. FAIR was initially directed by New York University's Yann LeCun, a deep learning Professor and Turing Award winner. Along with NYU's Center for Data Science, FAIR's initial goal was to research data science, machine learning and artificial intelligence. FAIR's goal was "to understand intelligence, to discover its fundamental principles, and to make machines significantly more intelligent". Research at FAIR pioneered the technology that led to face recognition and tagging in photographs and personalized feed recommendation. In 2014, Vladimir Vapnik joined FAIR. Vapnik is a pioneer in statistical learning, the co-inventor of the support-vector machine, and one of the developers of the Vapnik–Chervonenkis theory. FAIR opened a research center in Pairs, France in 2015, and subsequently launched smaller satellite research labs in Seattle, Pittsburgh, Tel Aviv, Montreal and London. In 2016, FAIR partnered with Google, Amazon, IBM and Microsoft in creating the Partnership on Artificial Intelligence to Benefit People and Society, an organization with focus on open licensed research, recommending best research practices and ethics, and discussing fairness, inclusivity, and transparency. In 2018, Jérôme Pesenti, former CTO of IBM's big data group, assumed the role of president of FAIR, while Yann LeCun stepped down to serve as chief AI scientist. In 2018, FAIR was placed 25th in the AI Research Rankings 2019, which ranked the top global organizations leading AI research. FAIR quickly rose to eighth position in 2019, and maintained eighth position in the 2020 rank. FAIR had approximately 200 staff in 2018, and had the goal to double that number by 2020.

FAIR's initial work included research in learning-model enabled memory networks, self-supervised learning and generative adversarial networks, text classification and translation, as well as computer vision. FAIR released Torch deep-learning modules and in 2017, FAIR released PyTorch, an open-source machine learning framework. PyTorch would then be used used in several deep learning technologies, such as Tesla's autopilot and Uber's Pyro. Also in 2017, FAIR discontinued a research project once AI bots developed a language that was unintelligible to humans, inciting conversations about dystopian fear of artificial intelligence going out of control. However, FAIR clarified that the research had been shut down due to efficiency, since they had accomplished their initial goal to understand how languages are generated, rather than out of fear. At the time, Yann LeChun criticized early media coverage of artificial intelligence saying that Terminator should stop being used as the cover of AI news articles.

FAIR was renamed MetaAI following the rebranding that changed Facebook, Inc to Meta Platforms Inc.

Current Research
In the February 23, 2022 live event Inside the Lab: Building for the Metaverse with AI, the Meta AI team discussed the major advancements in research and development in artificial intelligence. One such tool is the BuilderBot, which allows users to generate virtual worlds by using voice commands. Other tools include the No Language Left Behind, a system capable of automatic translation between written languages, and a Universal Speech Translator, a system capable of instantaneous speech-to-speech translation.

Computer Vision
Meta AI's computer vision research aims to extract information about the environment from digital images and videos. One example of technology towards scene understanding is panoptic segmentation, which recognizes objects in the foreground but also classifies the scenes in the background. Meta AI seeks to improve Visual Question Answering technology, in which a machine answers human user questions about images using cycle-consistency, having the machine generate a question in addition to the answer to address linguistic variations in the questions.

Natural Language Processing and Conversational AI
Artificial intelligence communication requires a machine to understand natural language and to generate language that is natural. Meta AI seeks to improve these technologies to improve safe communication regardless of what language the user might speak. Thus, a central task involves the generalization of natural language processing (NLP) technology to other languages. As such, Meta AI actively works on unsupervised machine translation. Meta AI seeks to improve natural-language interfaces by developing aspects of chitchat dialogue such as repetition, specificity, response-relatedness and question-asking, incorporating personality into image captioning , and generating creativity-based language.

In 2018, Meta AI launched the open-source PyText, a modeling framework focused on NLP systems.

Ranking & Recommendations
Meta AI research in ranking & recommendations is behind Facebook's News Feed, Instagram, and individually-recommended Ads and Search results. Meta AI introduced ReAgent, a toolset that generates decisions and evaluates user feedback

Systems Research
Machine learning and AI depend on the development of novel algorithms, software and hardware technologies. As such, Meta AI's systems research teams studies computer languages, compliers, and hardware applications.

Theory
Meta AI studies the mathematical and theoretical foundation to artificial intelligence through "techniques in learning theory, optimization, signal processing, and statistics"

Peer Review By Alyssa Guillu
Ranking and recommendations Meta AI research in ranking & recommendations is behind Facebook's News Feed, Instagram, and individually-recommended Ads and Search results. '''Ranking and recommendations is based on reasoning systems, a technology that aims to implement real-world decision-making. Reasoning systems rely on reinforcement learning, which in turn rely on performance feedback. These processes are difficult to automate. Meta Ai introduced ReAgent, an open source toolset that includes decision-making models, feedback generators and evaluators, and models to automate deployment of decision-making models to real-world applications. In addition to research and recommendations, these technologies are implemented in AI more broadly. For instance, the technologies in ReAgent have been used to teach robots how to learn to move autonomously and to teach machines how to play Go. ReAgent is an expansion of the Horizon program, the first open source reinforcement learning platform.' [Might be good to explain what the Horizon program is]''

Systems research
Machine learning and AI depend on the development of novel algorithms, software and hardware technologies. As such, Meta AI's systems research teams studies computer languages, compiler, and hardware applications. '''The systems research team addresses fundamental issues with computing such as floating-point error mitigation. The team is also responsible for evaluation of novel algorithm performance on state-of-the art machines. [' How big are these teams? Not crucial, but just a good point to add; what are the state of the art machines?]''

= Peer Review by Chad Berkich = All potential edits will be in italics and bolded to make them clearly standout. Comments are bolded and in brackets [example].

Meta AI is an artificial intelligence laboratory that belongs to Meta Platform Inc. (formerly known as Facebook, Inc. ). Meta AI seeks to develop artificial intelligence in the digital world, enhancing its augmented and artificial reality technologies. Meta AI is an academic research laboratory focused on generating knowledge for the broader AI community, and operates in parallel to Facebook's Applied Machine Learning (AML) team, which focuses on practical applications on its products. '''[These two sentences feel somewhat redundant. I know they don't convey the same information, but when reading them the starts felt very similar. I don't have a specific suggestion but I would recommend changing up the start on one of them or combining them to reduce that redundant feeling.]'''

History
Meta AI '''[Is it MetaAI or Meta AI? I added a space here because previously you have said its the latter.]''' started as Facebook Artificial Intelligence Research (FAIR) with locations in the Menlo Park, California headquarters, London, United Kingdom and a new laboratory in Manhattan. FAIR was officially announced in September, 2013. FAIR was initially directed by New York University's Yann LeCun, a deep learning Professor and Turing Award winner. Working with NYU's Center for Data Science, FAIR's initial goal was to research data science, machine learning, [note the comma added here] and artificial intelligence. FAIR's goal was "to understand intelligence, to discover its fundamental principles, and to make machines significantly more intelligent". Research at FAIR pioneered the technology that led to face recognition, [note the comma added here] tagging in photographs, [note the comma added here] and personalized feed recommendation. In 2014, Vladimir Vapnik joined FAIR. Vapnik is a pioneer in statistical learning, the co-inventor of the support-vector machine, and one of the developers of the Vapnik–Chervonenkis theory. FAIR opened a research center in Pairs, France in 2015, and subsequently launched smaller satellite research labs in Seattle, Pittsburgh, Tel Aviv, Montreal and London. In 2016, FAIR partnered with Google, Amazon, IBM, [note the comma added here] and Microsoft in creating the Partnership on Artificial Intelligence to Benefit People and Society, an organization with a focus on open licensed research, supporting ethical and effective research practices, and discussing fairness, inclusivity, and transparency. In 2018, Jérôme Pesenti, former CTO of IBM's big data group, assumed the role of president of FAIR, while [You can probably leave 'Yann' here but I would recommend cutting it] LeCun stepped down to serve as chief AI scientist. In 2018, FAIR was placed 25th in the AI Research Rankings 2019, which ranked the top global organizations leading AI research. FAIR quickly rose to eighth position in 2019, and maintained eighth position in the 2020 rank. FAIR had approximately 200 staff in 2018, and had the goal to double that number by 2020.

FAIR's initial work included research in learning-model enabled memory networks, self-supervised learning and generative adversarial networks, text classification and translation, as well as computer vision. FAIR released Torch deep-learning modules and in 2017, FAIR released PyTorch, an open-source machine learning framework. PyTorch was subsequently used [note that used here was put twice, and is now deleted] in several deep learning technologies, such as Tesla's autopilot and Uber's Pyro. Also in 2017, FAIR discontinued a research project once AI bots developed a language that was unintelligible to humans, inciting conversations about dystopian fear of artificial intelligence going out of control. However, FAIR clarified that the research had been shut down [You can probably leave 'due to efficiency,' here but I would recommend cutting it] since they had accomplished their initial goal to understand how languages are generated, rather than out of fear. At the time, [You can probably leave 'Yann' here but I would recommend cutting it] LeChun criticized early media coverage of artificial intelligence saying that Terminator should stop being used as the cover of AI news articles.

FAIR was renamed Meta AI [See earlier comment about spelling] following the rebranding that changed Facebook, Inc to Meta Platforms Inc.

Current Research
In the February 23, 2022 live event Inside the Lab: Building for the Metaverse with AI, the Meta AI team discussed the major advancements in research and development of artificial intelligence. One such tool is the BuilderBot, which allows users to generate virtual worlds by using voice commands. Other tools included the No Language Left Behind, a system capable of automatic translation between written languages, and a Universal Speech Translator, a system capable of instantaneous speech-to-speech translation.

Computer Vision
Meta AI's computer vision research aims to extract information about the environment '''[Is this environment like the natural world or what the computer is "seeing?" I would clarify here to be clear]''' from digital images and videos. One example of technology towards scene understanding [you haven't used this term before this, so I would suggest either explaining it or switching terms (I believe I understand what it is, but just for clarity I'm suggesting this) is panoptic segmentation, which recognizes objects in the foreground but also classifies the scenes in the background. Meta AI seeks to improve Visual Question Answering technology, in which a machine answers human user questions about images using cycle-consistency, having the machine generate a question in addition to the answer to address linguistic variations in the questions.

Natural Language Processing and Conversational AI
Artificial intelligence communication requires a machine to be able to understand and respond in natural language. Meta AI seeks to improve these technologies to improve safe communication regardless of what language the user might speak. Thus, a central task involves the generalization of natural language processing (NLP) technology to other languages. As such, Meta AI actively works on unsupervised machine translation. Meta AI seeks to improve natural-language interfaces by developing aspects of chitchat dialogue such as repetition, specificity, response-relatedness and question-asking, incorporating personality into image captioning , and generating creativity-based language.

In 2018, Meta AI launched the open-source PyText, a modeling framework focused on NLP systems.

Ranking & Recommendations
Meta AI research in ranking & recommendations is behind Facebook's News Feed, Instagram, and individually-recommended Ads and Search results. Meta AI introduced ReAgent, a toolset that generates decisions and evaluates user feedback '''. [note the period added here]'''

Systems Research
Machine learning and AI depend on the development of novel algorithms, software and hardware technologies. As such, Meta AI's systems research teams study computer languages, compliers, and hardware applications.

Theory
Meta AI studies the mathematical and theoretical foundation to artificial intelligence through "techniques in learning theory, optimization, signal processing, and statistics" '''. [note the period added here]'''

Overall Thoughts
This is a really great contribution. Only things suggested where grammar/spelling in a few places and some suggestions to streamline wording. Great job!

= Peer Review by Jack Casey =

Meta AI is an artificial intelligence laboratory that belongs to Meta Platform Inc. (formerly known as Facebook, Inc. ). Meta AI seeks to develop artificial intelligence in the digital world, enhancing its augmented and artificial reality technologies. Meta AI is an academic research laboratory focused on generating knowledge for the broader AI community, and operates in parallel to Facebook's Applied Machine Learning (AML) team, which focuses on practical applications on its products.

History
MetaAI started as Facebook Artificial Intelligence Research (FAIR) with locations in the Menlo Park, California headquarters, London, United Kingdom and a new laboratory in Manhattan. FAIR was officially announced in September, 2013. FAIR was initially directed by New York University's Yann LeCun, a deep learning Professor and Turing Award winner. Along with NYU's Center for Data Science, FAIR's initial goal was to research data science, machine learning and artificial intelligence. FAIR's goal was "to understand intelligence, to discover its fundamental principles, and to make machines significantly more intelligent". Research at FAIR pioneered the technology that led to face recognition and tagging in photographs and personalized feed recommendation. In 2014, Vladimir Vapnik joined FAIR. Vapnik is a pioneer in statistical learning, the co-inventor of the support-vector machine, and one of the developers of the Vapnik–Chervonenkis theory. FAIR opened a research center in Pairs, France in 2015, and subsequently launched smaller satellite research labs in Seattle, Pittsburgh, Tel Aviv, Montreal and London. In 2016, FAIR partnered with Google, Amazon, IBM and Microsoft in creating the Partnership on Artificial Intelligence to Benefit People and Society, an organization with focus on open licensed research, recommending best research practices and ethics, and discussing fairness, inclusivity, and transparency. In 2018, Jérôme Pesenti, former CTO of IBM's big data group, assumed the role of president of FAIR, while Yann LeCun stepped down to serve as chief AI scientist. In 2018, FAIR was placed 25th in the AI Research Rankings 2019, which ranked the top global organizations leading AI research. FAIR quickly rose to eighth position in 2019, and maintained eighth position in the 2020 rank. FAIR had approximately 200 staff in 2018, and had the goal to double that number by 2020. '[This is a fairly long, listy paragraph. I wonder if it should be broken up into several smaller paragraphs, or perhaps even a bulleted list of FAIR's accomplishments would read better. The content is good, though.]'

FAIR's initial work included research in learning-model enabled memory networks, self-supervised learning and generative adversarial networks, text classification and translation, as well as computer vision. FAIR released Torch deep-learning modules and in 2017, FAIR released PyTorch, an open-source machine learning framework. PyTorch would then be used used in several deep learning technologies, such as Tesla's autopilot and Uber's Pyro. Also in 2017, FAIR discontinued a research project once AI bots developed a language that was unintelligible to humans, inciting conversations about dystopian fear of artificial intelligence going out of control. However, FAIR clarified that the research had been shut down due to efficiency, since they had accomplished their initial goal to understand how languages are generated, rather than out of fear. At the time, Yann LeChun criticized early media coverage of artificial intelligence saying that Terminator should stop being used as the cover of AI news articles. '''[I like the first half of this paragraph, but the second half (the AI bots developing language) does not feel very relevant (it could/should be mentioned, but you go into quite a lot of depth). The Terminator bit, while fun, should probably be removed since it is not directly related to FAIR/MetaAI].'''

FAIR was renamed MetaAI following the rebranding that changed Facebook, Inc to Meta Platforms Inc.

Current Research
In the February 23, 2022 live event Inside the Lab: Building for the Metaverse with AI '[Quotes here, or Italics? It is a little hard to read without any formatting]', the Meta AI team discussed the major advancements in research and development in artificial intelligence. One such tool is the BuilderBot, which allows users to generate virtual worlds by using voice commands. Other tools include the No Language Left Behind, a system capable of automatic translation between written languages, and a Universal Speech Translator, a system capable of instantaneous speech-to-speech translation. [Should this paragraph go under a subheading?]

Computer Vision
Meta AI's computer vision research aims to extract information about the environment from digital images and videos. One example of technology towards scene understanding is panoptic segmentation, which recognizes objects in the foreground but also classifies the scenes in the background. Meta AI seeks to improve Visual Question Answering technology, in which a machine answers human user questions about images using cycle-consistency, having the machine generate a question in addition to the answer to address linguistic variations in the questions.

Natural Language Processing and Conversational AI
Artificial intelligence communication requires a machine to understand natural language and to generate language that is natural. Meta AI seeks to improve these technologies to improve safe communication regardless of what language the user might speak. Thus, a central task involves the generalization of natural language processing (NLP) technology to other languages. As such, Meta AI actively works on unsupervised machine translation. Meta AI seeks to improve natural-language interfaces by developing aspects of chitchat dialogue such as repetition, specificity, response-relatedness and question-asking, incorporating personality into image captioning , and generating creativity-based language.

In 2018, Meta AI launched the open-source PyText, a modeling framework focused on NLP systems.

Ranking & Recommendations
Meta AI research in ranking & recommendations is behind Facebook's News Feed, Instagram, and individually-recommended Ads and Search results. Meta AI introduced ReAgent, a toolset that generates decisions and evaluates user feedback

Systems Research
Machine learning and AI depend on the development of novel algorithms, software and hardware technologies. As such, Meta AI's systems research teams studies computer languages, compliers, and hardware applications.

Theory
Meta AI studies the mathematical and theoretical foundation to artificial intelligence through "techniques in learning theory, optimization, signal processing, and statistics" '[I am generally suspicious of using quotes in wikipedia articles, and would recommend a paraphrase instead. If you keep the quote, I think making it clear who is saying this in your text is important.]'

My Thoughts
'I think this is a good new article, and it is very very well sourced. I like the content, and I really only have thoughts on structure (which you can see above). I will also note that the last 3 subsections under Current Research are much shorter than the previous, and in an ideal world they could have a sentence or two added to each, but the reality is probably that sometimes there is just more to say about certain things.'

I like to conclude the peer reviews by reminding that I am liable to make the wrong judgements, so treat suggestions with a grain of salt.

Strong Work

-- JackCasey067 (talk) 20:29, 8 May 2022 (UTC)

Peer review by Matt
MetaAI started as Facebook Artificial Intelligence Research (FAIR) with locations in the Menlo Park, California headquarters, London, United Kingdom and a new laboratory in Manhattan. FAIR was officially announced in September, 2013. FAIR was initially directed by New York University's Yann LeCun, a deep learning Professor and Turing Award winner. Along with NYU's Center for Data Science, FAIR's initial goal was to research data science, machine learning and artificial intelligence. FAIR's goal was "to understand intelligence, to discover its fundamental principles, and to make machines significantly more intelligent". Research at FAIR pioneered the technology that led to face recognition and tagging in photographs and personalized feed recommendation. In 2014, Vladimir Vapnik joined FAIR. Vapnik is a pioneer in statistical learning, the co-inventor of the support-vector machine, and one of the developers of the Vapnik–Chervonenkis theory. FAIR opened a research center in Pairs, France in 2015, and subsequently launched smaller satellite research labs in Seattle, Pittsburgh, Tel Aviv, Montreal and London. In 2016, FAIR partnered with Google, Amazon, IBM and Microsoft in creating the Partnership on Artificial Intelligence to Benefit People and Society, an organization with focus on open licensed research, recommending best research practices and ethics, and discussing fairness, inclusivity, and transparency. In 2018, Jérôme Pesenti, former CTO of IBM's big data group, assumed the role of president of FAIR, while Yann LeCun stepped down to serve as chief AI scientist. In 2018, FAIR was placed 25th in the AI Research Rankings 2019, which ranked the top global organizations leading AI research. FAIR quickly rose to eighth position in 2019, and maintained eighth position in the 2020 rank. FAIR had approximately 200 staff in 2018, and had the goal to double that number by 2020. I am not sure how much does factual info help with this, i think it would be, however, extremly helpful if you can follow up each new development with their effects on the algorithm/project or research.

FAIR's initial work included research in learning-model enabled memory networks, self-supervised learning and generative adversarial networks, text classification and translation, as well as computer vision. FAIR released Torch deep-learning modules and in 2017, FAIR released PyTorch, an open-source machine learning framework. PyTorch would then be used used in several deep learning technologies, such as Tesla's autopilot  and Uber's Pyro. Also in 2017, FAIR discontinued a research project once AI bots developed a language that was unintelligible to humans, inciting conversations about dystopian fear of artificial intelligence going out of control. However, FAIR clarified that the research had been shut down due to efficiency, since they had accomplished their initial goal to understand how languages are generated, rather than out of fear. At the time, Yann LeChun criticized early media coverage of artificial intelligence saying that Terminator should stop being used as the cover of AI news articles. It would be nice if this part is conneted with the previous paragraph so reader can get s sense of time line.

FAIR was renamed MetaAI following the rebranding that changed Facebook, Inc to Meta Platforms Inc.

Current Research[edit]
In the February 23, 2022 live event Inside the Lab: Building for the Metaverse with AI, the Meta AI team discussed the major advancements in research and development in artificial intelligence. One such tool is the BuilderBot, which allows users to generate virtual worlds by using voice commands. Other tools include the No Language Left Behind, a system capable of automatic translation between written languages, and a Universal Speech Translator, a system capable of instantaneous speech-to-speech translation.

Computer Vision[edit]
Meta AI's computer vision research aims to extract information about the environment from digital images and videos. One example of technology towards scene understanding is panoptic segmentation, which recognizes objects in the foreground but also classifies the scenes in the background. Meta AI seeks to improve Visual Question Answering technology, in which a machine answers human user questions about images using cycle-consistency, having the machine generate a question in addition to the answer to address linguistic variations in the questions.

Natural Language Processing and Conversational AI[edit]
Artificial intelligence communication requires a machine to understand natural language and to generate language that is natural. Meta AI seeks to improve these technologies to improve safe communication regardless of what language the user might speak. Thus, a central task involves the generalization of natural language processing (NLP) technology to other languages. As such, Meta AI actively works on unsupervised machine translation. Meta AI seeks to improve natural-language interfaces by developing aspects of chitchat dialogue such as repetition, specificity, response-relatedness and question-asking, incorporating personality into image captioning, and generating creativity-based language.

In 2018, Meta AI launched the open-source PyText, a modeling framework focused on NLP systems.

Ranking & Recommendations[edit]
Meta AI research in ranking & recommendations is behind Facebook's News Feed, Instagram, and individually-recommended Ads and Search results. Meta AI introduced ReAgent, a toolset that generates decisions and evaluates user feedback

Systems Research[edit]
Machine learning and AI depend on the development of novel algorithms, software and hardware technologies. As such, Meta AI's systems research teams studies computer languages, compliers, and hardware applications.

Theory[edit]
Meta AI studies the mathematical and theoretical foundation to artificial intelligence through "techniques in learning theory, optimization, signal processing, and statistics"'

'''The last 3 research fields seems lacking in contnet. I think Facebook definitely have lots of ranking and recommendations algo research, I also wonder if you should add/link pytorch to this page'''