DOCSLIB.ORG

Machine Learning for Speech Recognition by Alice Coucke, Head of Machine Learning Research

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Machine Learning for Speech Recognition by Alice Coucke, Head of Machine Learning Research Machine Learning for Speech Recognition by Alice Coucke, Head of Machine Learning Research @alicecoucke [email protected] Outline: 1. Recent advances in machine learning 2. From physics to machine learning IA en général Speech & NLP Startup 3. Working at Snips (now Sonos) Snips A lot of applications in the real world, quite a difference from theoretical work Un des rares domaines scientifiques où la théorie et la pratique sont très emmêlées Recent Advances in Applied Machine Learning A lot of applications in the real world, quite a difference from theoretical work Reinforcement learning Learning goal-oriented behavior within simulated environments Go Starcraft II Dota 2 AlphaGo (Deepmind, 2016) AlphaStar (Deepmind) OpenAI Five (OpenAI) Play-driven learning for robots Sim-to-real dexterity learning (Google Brain) Project BLUE (UC Berkeley) Machine Learning for Life Sciences Deep learning applied to biology and medicine Protein folding & structure Cardiac arrhythmia prediction Eye disease diagnosis prediction (NHS, UCL, Deepmind) from ECGs AlphaFold (Deepmind) (Stanford) Reconstruct speech from neural activity Limb control restoration (UCSF) (Batelle, Ohio State Univ) Computer vision High-level understanding of digital images or videos GANs for image generation (Heriot Watt Univ, DeepMind) « Common sense » understanding of actions in videos (TwentyBn, DeepMind, MIT, IBM…) GANs for artificial video dubbing GAN for full body synthesis (Synthesia) (DataGrid) From physics to machine learning and back A surge of interest from the physics community NeurIPS 2019: workshop on « machine learning and the physical sciences » Speech and language Understand and analyze human speech { intent: FindWeather, entities: { datetime: 11/28/2019, location: Paris } } Speech transcription Spoken language Text & speech generation Human Parity (Microsoft) understanding GPT-2 (Open AI) (Super)GLUE benchmarks (Google, Bert (Google) Facebook, IBM, Stanford …) XLNet (CMU) … Speech and language Understand and analyze human speech { intent: FindWeather, entities: { datetime: 11/28/2019, location: Paris } } Speech transcription Spoken language Text & speech generation Human Parity (Microsoft) understanding GPT-2 (Open AI) (Super)GLUE benchmarks (Google, Bert (Google) Facebook, IBM, Stanford …) XLNet (CMU) … � � Neural machine translation Voice activity detection Speaker identification Sentiment analysis Detect emotions in text and Unsupervised MT (Facebook) Detect speech from audio Recognize unique speakers speech Fairness, ethics, and explainability We, as scientists, have a say in the future of AI From physics to machine learning Décrire mon parcours My background Parler de la thèse From physics to machine learning Parler du post doc Pourquoi les physiciens / chercheurs sont une bonne formation - appréhender des théories complexes et les appliquer à des domaines variés - Programmation skills - Formation par la recherche : communiquer des résultats, lire un état de l’art, écrire des articles - Experimental method Décrire mon parcours My background Parler de la thèse From physics to machine learning Parler du post doc Pourquoi les physiciens / chercheurs 2012: M2 ICFP Theoretical physics sont une bonne formation - appréhender des théories complexes et les appliquer à des 2013-2016: PhD in statistical physics @LPTENS domaines variés - Programmation skills - Formation par la recherche : communiquer des résultats, lire un état de l’art, écrire des articles - Experimental method Décrire mon parcours My background Parler de la thèse From physics to machine learning Parler du post doc Pourquoi les physiciens / chercheurs 2012: M2 ICFP Theoretical physics sont une bonne formation - appréhender des théories complexes et les appliquer à des 2013-2016: PhD in statistical physics @LPTENS domaines variés - Programmation skills - Formation par la recherche : communiquer des résultats, lire un état de l’art, écrire des articles - Experimental method Décrire mon parcours My background Parler de la thèse From physics to machine learning Parler du post doc Pourquoi les physiciens / chercheurs 2012: M2 ICFP Theoretical physics sont une bonne formation - appréhender des théories complexes et les appliquer à des 2013-2016: PhD in statistical physics @LPTENS domaines variés - Programmation skills - Formation par la recherche : communiquer des résultats, lire un état de l’art, écrire des articles - Experimental method Feb 2017: senior ML scientist @ Snips 2019: director of ML research @ Snips Today: head of ML research @ Sonos, Inc. A few takeaways (please go ask other people too) A few takeaways (please go ask other people too) • PhD? Postdoc? • Working at a startup company • Physicists and machine learning A few takeaways (please go ask other people too) • PhD? Postdoc? • Working at a startup company • Physicists and machine learning Physicists @ Snips A. C. Raffaele Tavarone Francesco Caltagirone Alaa Saade Stéphane d’Ascoli Sr ML Scientist Sr ML Scientist Sr ML Scientist ML research intern Acoustics team Tech Lead Now: DeepMind Now: PhD ENS & FAIR Language team A few takeaways (please go ask other people too) • PhD? Postdoc? • Working at a startup company • Physicists and machine learning Physicists @ Snips A. C. Raffaele Tavarone Francesco Caltagirone Alaa Saade Stéphane d’Ascoli Sr ML Scientist Sr ML Scientist Sr ML Scientist ML research intern Acoustics team Tech Lead Now: DeepMind Now: PhD ENS & FAIR Language team Machine Learning at Snips (now Sonos) ML à Snips comme on peut faire du ML et de la science en dehors de l’académique Speech & language recognition - Privacy - pas cloud données en local —> contraintes en + techniques - resources computaitonelles - Peu de données pour entraîner des données - Chaque bloc est optimisé adapté Spoken Language Understanding From speech to meaning for voice assistants Automatic Speech Recognition Engine Language modeling Natural Intent: t ɜ r n ɑ n ð ə Acoustic Language Turn on the Language SwitchLightOn l a ɪ t s ɪ n ð ə model model lights in the Understanding ˈl ɪ v ɪ ŋ r u m living room Engine Slots: room: living room Wake Word ACR ICF Dialogue Custom Wake Automatic Command Intent Classification Multiturn Dialogue Word Recognition and Filling and Response Audio Frontend Action Code Spoken Language Understanding From speech to meaning for voice assistants Automatic Speech Recognition Engine Language modeling Natural Intent: t ɜ r n ɑ n ð ə Acoustic Language Turn on the Language SwitchLightOn l a ɪ t s ɪ n ð ə model model lights in the Understanding ˈl ɪ v ɪ ŋ r u m living room Engine Slots: room: living room Spoken Language Understanding From speech to meaning for voice assistants Automatic Speech Recognition Engine Language modeling Natural Intent: t ɜ r n ɑ n ð ə Acoustic Language Turn on the Language SwitchLightOn l a ɪ t s ɪ n ð ə model model lights in the Understanding ˈl ɪ v ɪ ŋ r u m living room Engine Slots: room: living room Deep neural network Proba over phones /a/ /b/ /c/ /d/ /e/ time Spoken Language Understanding From speech to meaning for voice assistants Automatic Speech Recognition Engine Language modeling Natural Intent: t ɜ r n ɑ n ð ə Acoustic Language Turn on the Language SwitchLightOn l a ɪ t s ɪ n ð ə Language model model lights in the Understanding ˈl ɪ v ɪ ŋ r u m model living room Engine Slots: room: living room Proba over phones Decoding graph Turn on the lights in the living room 0.85 Turn off the lights in the living room 0.75 Set the lights in the living room 0.60 /a/ /b/ /c/ /d/ /e/ time Spoken Language Understanding From speech to meaning for voice assistants Automatic Speech Recognition Engine Language modeling Natural Intent: t ɜ r n ɑ n ð ə Acoustic Language Turn on the Language SwitchLightOn l a ɪ t s ɪ n ð ə model model lights in the Understanding ˈl ɪ v ɪ ŋ r u m living room Engine Slots: room: living room Logistic regression Conditional Random Field Turn on the lights Intent: Slots: in the living room SwitchLightOn room: living room Spoken Language Understanding Offline & on device Automatic Speech Recognition Engine Language modeling Natural Intent: t ɜ r n ɑ n ð ə Acoustic Language Turn on the Language SwitchLightOn l a ɪ t s ɪ n ð ə model model lights in the Understanding ˈl ɪ v ɪ ŋ r u m living room Engine Slots: room: living room ❌ ❌ ✅ Wake Word ACR ICF Dialogue Custom Wake Automatic Command Intent Classification Multiturn Dialogue Word Recognition and Filling and Response Audio Frontend Action Code Our approach Our own voice in a vast ecosystem Privacy by design Resource constrained ML: small data & hardware ✓ A new popular trend in the ML community (low resource ML, transfer learning, miniaturization, etc) ✓ Numerous conferences and workshops on the topic ✓ Towards a safer, greener and more private conversational AI Research activity Publishing research in industry Snips voice platform: Efficient keyword spotting an embedded spoken language using dilated convolutions Federated learning for Spoken language understanding system for and gating keyword spotting understanding on the edge private-by-design voice interfaces arXiv arXiv arXiv arXiv There is a lot to do in research in the industry, with us joining Sonos we expect to grow the team and doing even more exploratory stuff. ICML 2018 ICASSP 2019 ICASSP 2019 NeurIPS 2019 workshop PiMLAI Main track Main track Workshop EMC2 Cited 35 times to date ✓ Publish open & reproducible benchmarks: ‣ ~200 access granted to researchers to our open speech datasets ‣ Snips dataset for NLU is the new academic standard Thank you for your attention Questions? @alicecoucke [email protected].
Load more

Recommended publications
  • Aviation Speech Recognition System Using Artificial Intelligence
    SPEECH RECOGNITION SYSTEM OVERVIEW AVIATION SPEECH RECOGNITION SYSTEM USING ARTIFICIAL INTELLIGENCE Appareo’s embeddable AI model for air traffic control (ATC) transcription makes the company one of the world leaders in aviation artificial intelligence. TECHNOLOGY APPLICATION As featured in Appareo’s two flight apps for pilots, Stratus InsightTM and Stratus Horizon Pro, the custom speech recognition system ATC Transcription makes the company one of the world leaders in aviation artificial intelligence. ATC Transcription is a recurrent neural network that transcribes analog or digital aviation audio into text in near-real time. This in-house artificial intelligence, trained with Appareo’s proprietary dataset of flight-deck audio, takes the terabytes of training data and its accompanying transcriptions and reduces that content to a powerful 160 MB model that can be run inside the aircraft. BENEFITS OF ATC TRANSCRIPTION: • Provides improved situational awareness by identifying, capturing, and presenting ATC communications relevant to your aircraft’s operation for review or replay. • Provides the ability to stream transcribed speech to on-board (avionic) or off-board (iPad/ iPhone) display sources. • Provides a continuous representation of secondary audio channels (ATIS, AWOS, etc.) for review in textual form, allowing your attention to focus on a single channel. • ...and more. What makes ATC Transcription special is the context-specific training work that Appareo has done, using state-of-the-art artificial intelligence development techniques, to create an AI that understands the aviation industry vocabulary. Other natural language processing work is easily confused by the cadence, noise, and vocabulary of the aviation industry, yet Appareo’s groundbreaking work has overcome these barriers and brought functional artificial intelligence to the cockpit.
    [Show full text]
  • Artificial Intelligence: with Great Power Comes Great Responsibility
    ARTIFICIAL INTELLIGENCE: WITH GREAT POWER COMES GREAT RESPONSIBILITY JOINT HEARING BEFORE THE SUBCOMMITTEE ON RESEARCH AND TECHNOLOGY & SUBCOMMITTEE ON ENERGY COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY HOUSE OF REPRESENTATIVES ONE HUNDRED FIFTEENTH CONGRESS SECOND SESSION JUNE 26, 2018 Serial No. 115–67 Printed for the use of the Committee on Science, Space, and Technology ( Available via the World Wide Web: http://science.house.gov U.S. GOVERNMENT PUBLISHING OFFICE 30–877PDF WASHINGTON : 2018 COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY HON. LAMAR S. SMITH, Texas, Chair FRANK D. LUCAS, Oklahoma EDDIE BERNICE JOHNSON, Texas DANA ROHRABACHER, California ZOE LOFGREN, California MO BROOKS, Alabama DANIEL LIPINSKI, Illinois RANDY HULTGREN, Illinois SUZANNE BONAMICI, Oregon BILL POSEY, Florida AMI BERA, California THOMAS MASSIE, Kentucky ELIZABETH H. ESTY, Connecticut RANDY K. WEBER, Texas MARC A. VEASEY, Texas STEPHEN KNIGHT, California DONALD S. BEYER, JR., Virginia BRIAN BABIN, Texas JACKY ROSEN, Nevada BARBARA COMSTOCK, Virginia CONOR LAMB, Pennsylvania BARRY LOUDERMILK, Georgia JERRY MCNERNEY, California RALPH LEE ABRAHAM, Louisiana ED PERLMUTTER, Colorado GARY PALMER, Alabama PAUL TONKO, New York DANIEL WEBSTER, Florida BILL FOSTER, Illinois ANDY BIGGS, Arizona MARK TAKANO, California ROGER W. MARSHALL, Kansas COLLEEN HANABUSA, Hawaii NEAL P. DUNN, Florida CHARLIE CRIST, Florida CLAY HIGGINS, Louisiana RALPH NORMAN, South Carolina DEBBIE LESKO, Arizona SUBCOMMITTEE ON RESEARCH AND TECHNOLOGY HON. BARBARA COMSTOCK, Virginia, Chair FRANK D. LUCAS, Oklahoma DANIEL LIPINSKI, Illinois RANDY HULTGREN, Illinois ELIZABETH H. ESTY, Connecticut STEPHEN KNIGHT, California JACKY ROSEN, Nevada BARRY LOUDERMILK, Georgia SUZANNE BONAMICI, Oregon DANIEL WEBSTER, Florida AMI BERA, California ROGER W. MARSHALL, Kansas DONALD S. BEYER, JR., Virginia DEBBIE LESKO, Arizona EDDIE BERNICE JOHNSON, Texas LAMAR S.
    [Show full text]
  • A Comparison of Online Automatic Speech Recognition Systems and the Nonverbal Responses to Unintelligible Speech
    A Comparison of Online Automatic Speech Recognition Systems and the Nonverbal Responses to Unintelligible Speech Joshua Y. Kim1, Chunfeng Liu1, Rafael A. Calvo1*, Kathryn McCabe2, Silas C. R. Taylor3, Björn W. Schuller4, Kaihang Wu1 1 University of Sydney, Faculty of Engineering and Information Technologies 2 University of California, Davis, Psychiatry and Behavioral Sciences 3 University of New South Wales, Faculty of Medicine 4 Imperial College London, Department of Computing Abstract large-scale commercial products such as Google Home and Amazon Alexa. Mainstream ASR Automatic Speech Recognition (ASR) systems use only voice as inputs, but there is systems have proliferated over the recent potential benefit in using multi-modal data in order years to the point that free platforms such to improve accuracy [1]. Compared to machines, as YouTube now provide speech recognition services. Given the wide humans are highly skilled in utilizing such selection of ASR systems, we contribute to unstructured multi-modal information. For the field of automatic speech recognition example, a human speaker is attuned to nonverbal by comparing the relative performance of behavior signals and actively looks for these non- two sets of manual transcriptions and five verbal ‘hints’ that a listener understands the speech sets of automatic transcriptions (Google content, and if not, they adjust their speech Cloud, IBM Watson, Microsoft Azure, accordingly. Therefore, understanding nonverbal Trint, and YouTube) to help researchers to responses to unintelligible speech can both select accurate transcription services. In improve future ASR systems to mark uncertain addition, we identify nonverbal behaviors transcriptions, and also help to provide feedback so that are associated with unintelligible speech, as indicated by high word error that the speaker can improve his or her verbal rates.
    [Show full text]
  • Learned in Speech Recognition: Contextual Acoustic Word Embeddings
    LEARNED IN SPEECH RECOGNITION: CONTEXTUAL ACOUSTIC WORD EMBEDDINGS Shruti Palaskar∗, Vikas Raunak∗ and Florian Metze Carnegie Mellon University, Pittsburgh, PA, U.S.A. fspalaska j vraunak j fmetze [email protected] ABSTRACT model [10, 11, 12] trained for direct Acoustic-to-Word (A2W) speech recognition [13]. Using this model, we jointly learn to End-to-end acoustic-to-word speech recognition models have re- automatically segment and classify input speech into individual cently gained popularity because they are easy to train, scale well to words, hence getting rid of the problem of chunking or requiring large amounts of training data, and do not require a lexicon. In addi- pre-defined word boundaries. As our A2W model is trained at the tion, word models may also be easier to integrate with downstream utterance level, we show that we can not only learn acoustic word tasks such as spoken language understanding, because inference embeddings, but also learn them in the proper context of their con- (search) is much simplified compared to phoneme, character or any taining sentence. We also evaluate our contextual acoustic word other sort of sub-word units. In this paper, we describe methods embeddings on a spoken language understanding task, demonstrat- to construct contextual acoustic word embeddings directly from a ing that they can be useful in non-transcription downstream tasks. supervised sequence-to-sequence acoustic-to-word speech recog- Our main contributions in this paper are the following: nition model using the learned attention distribution. On a suite 1. We demonstrate the usability of attention not only for aligning of 16 standard sentence evaluation tasks, our embeddings show words to acoustic frames without any forced alignment but also for competitive performance against a word2vec model trained on the constructing Contextual Acoustic Word Embeddings (CAWE).
    [Show full text]
  • Synthesis and Recognition of Speech Creating and Listening to Speech
    ISSN 1883-1974 (Print) ISSN 1884-0787 (Online) National Institute of Informatics News NII Interview 51 A Combination of Speech Synthesis and Speech Oct. 2014 Recognition Creates an Affluent Society NII Special 1 “Statistical Speech Synthesis” Technology with a Rapidly Growing Application Area NII Special 2 Finding Practical Application for Speech Recognition Feature Synthesis and Recognition of Speech Creating and Listening to Speech A digital book version of “NII Today” is now available. http://www.nii.ac.jp/about/publication/today/ This English language edition NII Today corresponds to No. 65 of the Japanese edition [Advance Notice] Great news! NII Interview Yamagishi-sensei will create my voice! Yamagishi One result is a speech translation sys- sound. Bit (NII Character) A Combination of Speech Synthesis tem. This system recognizes speech and translates it Ohkawara I would like your comments on the fu- using machine translation to synthesize speech, also ture challenges. and Speech Recognition automatically translating it into every language to Ono The challenge for speech recognition is how speak. Moreover, the speech is created with a human close it will come to humans in the distant speech A Word from the Interviewer voice. In second language learning, you can under- case. If this study is advanced, it will be possible to Creates an Affluent Society stand how you should pronounce it with your own summarize the contents of a meeting and to automati- voice. If this is further advanced, the system could cally take the minutes. If a robot understands the con- have an actor in a movie speak in a different language tents of conversations by multiple people in a natural More and more people have begun to use smart- a smartphone, I use speech input more often.
    [Show full text]
  • Natural Language Processing in Speech Understanding Systems
    Working Paper Series ISSN 11 70-487X Natural language processing in speech understanding systems by Dr Geoffrey Holmes Working Paper 92/6 September, 1992 © 1992 by Dr Geoffrey Holmes Department of Computer Science The University of Waikato Private Bag 3105 Hamilton, New Zealand Natural Language P rocessing In Speech Understanding Systems G. Holmes Department of Computer Science, University of Waikato, New Zealand Overview Speech understanding systems (SUS's) came of age in late 1071 as a result of a five year devel­ opment. programme instigated by the Information Processing Technology Office of the Advanced Research Projects Agency (ARPA) of the Department of Defense in the United States. The aim of the progranune was to research and tlevelop practical man-machine conuuunication system s. It has been argued since, t hat. t he main contribution of this project was not in the development of speech science, but in the development of artificial intelligence. That debate is beyond the scope of th.is paper, though no one would question the fact. that the field to benefit most within artificial intelligence as a result of this progranune is natural language understan ding. More recent projects of a similar nature, such as projects in the Unite<l Kiug<lom's ALVEY programme and Ew·ope's ESPRIT programme have added further developments to this important field. Th.is paper presents a review of some of the natural language processing techniques used within speech understanding syst:ems. In particular. t.ecl.miq11es for handling syntactic, semantic and pragmatic informat.ion are ,Uscussed. They are integrated into SUS's as knowledge sources.
    [Show full text]
  • Arxiv:2007.00183V2 [Eess.AS] 24 Nov 2020
    WHOLE-WORD SEGMENTAL SPEECH RECOGNITION WITH ACOUSTIC WORD EMBEDDINGS Bowen Shi, Shane Settle, Karen Livescu TTI-Chicago, USA fbshi,settle.shane,[email protected] ABSTRACT Segmental models are sequence prediction models in which scores of hypotheses are based on entire variable-length seg- ments of frames. We consider segmental models for whole- word (“acoustic-to-word”) speech recognition, with the feature vectors defined using vector embeddings of segments. Such models are computationally challenging as the number of paths is proportional to the vocabulary size, which can be orders of magnitude larger than when using subword units like phones. We describe an efficient approach for end-to-end whole-word segmental models, with forward-backward and Viterbi de- coding performed on a GPU and a simple segment scoring function that reduces space complexity. In addition, we inves- tigate the use of pre-training via jointly trained acoustic word embeddings (AWEs) and acoustically grounded word embed- dings (AGWEs) of written word labels. We find that word error rate can be reduced by a large margin by pre-training the acoustic segment representation with AWEs, and additional Fig. 1. Whole-word segmental model for speech recognition. (smaller) gains can be obtained by pre-training the word pre- Note: boundary frames are not shared. diction layer with AGWEs. Our final models improve over segmental models, where the sequence probability is com- prior A2W models. puted based on segment scores instead of frame probabilities. Index Terms— speech recognition, segmental model, Segmental models have a long history in speech recognition acoustic-to-word, acoustic word embeddings, pre-training research, but they have been used primarily for phonetic recog- nition or as phone-level acoustic models [11–18].
    [Show full text]
  • Improving Named Entity Recognition Using Deep Learning with Human in the Loop
    Demonstration Improving Named Entity Recognition using Deep Learning with Human in the Loop Ticiana L. Coelho da Silva Regis Pires Magalhães José Antônio F. de Macêdo Insight Data Science Lab Insight Data Science Lab Insight Data Science Lab Fortaleza, Ceara, Brazil Fortaleza, Ceara, Brazil Fortaleza, Ceara, Brazil [email protected] [email protected] [email protected] David Araújo Abreu Natanael da Silva Araújo Vinicius Teixeira de Melo Insight Data Science Lab Insight Data Science Lab Insight Data Science Lab Fortaleza, Ceara, Brazil Fortaleza, Ceara, Brazil Fortaleza, Ceara, Brazil [email protected] [email protected] [email protected] Pedro Olímpio Pinheiro Paulo A. L. Rego Aloisio Vieira Lira Neto Insight Data Science Lab Federal University of Ceara Brazilian Federal Highway Police Fortaleza, Ceara, Brazil Fortaleza, Ceara, Brazil Fortaleza, Ceara, Brazil [email protected] [email protected] [email protected] ABSTRACT Even, orthographic features and language-specific knowledge Named Entity Recognition (NER) is a challenging problem in Nat- resources as gazetteers are widely used in NER, such approaches ural Language Processing (NLP). Deep Learning techniques have are costly to develop, especially for new languages and new do- been extensively applied in NER tasks because they require little mains, making NER a challenge to adapt to these scenarios[10]. feature engineering and are free from language-specific resources, Deep Learning models have been extensively used in NER learning important features from word or character embeddings tasks [3, 5, 9] because they require little feature engineering and trained on large amounts of data. However, these techniques they may learn important features from word or character embed- are data-hungry and require a massive amount of training data.
    [Show full text]
  • AI for Health Examples from Industry, Government, and Academia Table of Contents
    guidehouse.com AI for Health Examples from industry, government, and academia Table of Contents Abstract 3 Introduction to artificial intelligence 4 Introduction to health data 7 Molecules: AI for understanding biology and developing therapies 10 Fundamental research 10 Drug development 11 Patients: AI for predicting disease and improving care 13 Diagnostics and outcome risks 13 Personalized medicine 14 Groups: AI for optimizing clinical trials and safeguarding public health 15 Clinical trials 15 Public health 17 Conclusion and outlook 18 2 Guidehouse Abstract This paper is directed toward a health-informed reader who is curious about the developments and potential of artificial intelligence (AI) in the health space, but could equally be read by AI practitioners curious about how their knowledge and methods are being used to advance human health. We present a brief, equation-free introduction to AI and its major subfields in order to provide a framework for understanding the technical context of the examples that follow. We discuss the various data sources available for questions of health and life sciences, as well as the unique challenges inherent to these fields. We then consider recent (past five years) applications of AI that have already had tangible, measurable impact to the advancement of biomedical knowledge and the development of new and improved treatments. These examples are organized by scale, ranging from the molecule (fundamental research and drug development) to the patient (diagnostics, risk-scoring, and personalized medicine) to the group (clinical trials and public health). Finally, we conclude with a brief summary and our outlook for the future of AI for health.
    [Show full text]
  • Pre-Training of Deep Bidirectional Protein Sequence Representations with Structural Information
    Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000. Digital Object Identifier 10.1109/ACCESS.2017.DOI Pre-training of Deep Bidirectional Protein Sequence Representations with Structural Information SEONWOO MIN1,2, SEUNGHYUN PARK3, SIWON KIM1, HYUN-SOO CHOI4, BYUNGHAN LEE5 (Member, IEEE), and SUNGROH YOON1,6 (Senior Member, IEEE) 1Department of Electrical and Computer engineering, Seoul National University, Seoul 08826, South Korea 2LG AI Research, Seoul 07796, South Korea 3Clova AI Research, NAVER Corp., Seongnam 13561, South Korea 4Department of Computer Science and Engineering, Kangwon National University, Chuncheon 24341, South Korea 5Department of Electronic and IT Media Engineering, Seoul National University of Science and Technology, Seoul 01811, South Korea 6Interdisciplinary Program in Artificial Intelligence, ASRI, INMC, and Institute of Engineering Research, Seoul National University, Seoul 08826, South Korea Corresponding author: Byunghan Lee ([email protected]) or Sungroh Yoon ([email protected]) This research was supported by the National Research Foundation (NRF) of Korea grants funded by the Ministry of Science and ICT (2018R1A2B3001628 (S.Y.), 2014M3C9A3063541 (S.Y.), 2019R1G1A1003253 (B.L.)), the Ministry of Agriculture, Food and Rural Affairs (918013-4 (S.Y.)), and the Brain Korea 21 Plus Project in 2021 (S.Y.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. ABSTRACT Bridging the exponentially growing gap between the numbers of unlabeled and labeled protein sequences, several studies adopted semi-supervised learning for protein sequence modeling. In these studies, models were pre-trained with a substantial amount of unlabeled data, and the representations were transferred to various downstream tasks.
    [Show full text]
  • Biological Structure and Function Emerge from Scaling Unsupervised Learning to 250 Million Protein Sequences
    bioRxiv preprint doi: https://doi.org/10.1101/622803; this version posted May 29, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. BIOLOGICAL STRUCTURE AND FUNCTION EMERGE FROM SCALING UNSUPERVISED LEARNING TO 250 MILLION PROTEIN SEQUENCES Alexander Rives ∗y z Siddharth Goyal ∗x Joshua Meier ∗x Demi Guo ∗x Myle Ott x C. Lawrence Zitnick x Jerry Ma y x Rob Fergus y z x ABSTRACT In the field of artificial intelligence, a combination of scale in data and model capacity enabled by unsupervised learning has led to major advances in representation learning and statistical generation. In biology, the anticipated growth of sequencing promises unprecedented data on natural sequence diversity. Learning the natural distribution of evolutionary protein sequence variation is a logical step toward predictive and generative modeling for biology. To this end we use unsupervised learning to train a deep contextual language model on 86 billion amino acids across 250 million sequences spanning evolutionary diversity. The resulting model maps raw sequences to representations of biological properties without labels or prior domain knowledge. The learned representation space organizes sequences at multiple levels of biological granularity from the biochemical to proteomic levels. Unsupervised learning recovers information about protein structure: secondary structure and residue-residue contacts can be identified by linear projections from the learned representations. Training language models on full sequence diversity rather than individual protein families increases recoverable information about secondary structure.
    [Show full text]
  • Towards Incremental Agent Enhancement for Evolving Games
    Evaluating Reinforcement Learning Algorithms For Evolving Military Games James Chao*, Jonathan Sato*, Crisrael Lucero, Doug S. Lange Naval Information Warfare Center Pacific *Equal Contribution ffi[email protected] Abstract games in 2013 (Mnih et al. 2013), Google DeepMind devel- oped AlphaGo (Silver et al. 2016) that defeated world cham- In this paper, we evaluate reinforcement learning algorithms pion Lee Sedol in the game of Go using supervised learning for military board games. Currently, machine learning ap- and reinforcement learning. One year later, AlphaGo Zero proaches to most games assume certain aspects of the game (Silver et al. 2017b) was able to defeat AlphaGo with no remain static. This methodology results in a lack of algorithm robustness and a drastic drop in performance upon chang- human knowledge and pure reinforcement learning. Soon ing in-game mechanics. To this end, we will evaluate general after, AlphaZero (Silver et al. 2017a) generalized AlphaGo game playing (Diego Perez-Liebana 2018) AI algorithms on Zero to be able to play more games including Chess, Shogi, evolving military games. and Go, creating a more generalized AI to apply to differ- ent problems. In 2018, OpenAI Five used five Long Short- term Memory (Hochreiter and Schmidhuber 1997) neural Introduction networks and a Proximal Policy Optimization (Schulman et al. 2017) method to defeat a professional DotA team, each AlphaZero (Silver et al. 2017a) described an approach that LSTM acting as a player in a team to collaborate and achieve trained an AI agent through self-play to achieve super- a common goal. AlphaStar used a transformer (Vaswani et human performance.
    [Show full text]