Natasha Jaques's guest appearances

Natasha Jaques

TalkRL: The Reinforcement Learning Podcast

You Belong in AI!

Kamyar Azizzadenesheli

Scott Fujimoto

Jessica Hamrick

Samuel Castro

Ashley Hill

Michael Littman

Rohin Shah

Robert Osazuwa Ness

Jordan Terry

Amy Zhang

Michael Dennis

Jess Whittlestone

Taylor Killian

Ben Eysenbach

Csaba Szepesvari

OwlTail

This episode features Natasha Jaques (she/her), a Senior Research Scientist at Google Brain! Natasha received her PhD in Computer Science at MIT in 2019, and she was a part of various research groups like DeepMind. In this episode, Natasha shares about her background, discusses representation in AI, and gives valuable life advice. Natasha will be joining the University of Washington School of Computer Science as an assistant professor starting in January 2024.

This episode features Natasha Jaques (she/her), a Senior Research Scientist at Google Brain! Natasha received her PhD in Computer Science at MIT in 2019, and she was a part of various research groups like DeepMind. In this episode, Natasha shares about her background, discusses representation in AI, and gives valuable life advice. Natasha will be joining the University of Washington School of Computer Science as an assistant professor starting in January 2024.

You Belong in AI! S3 E3 Natasha Jaques

Hear about why OpenAI cites her work in RLHF and dialog models, approaches to rewards in RLHF, ChatGPT, Industry vs Academia, PsiPhi-Learning, AGI and more!  Dr Natasha Jaques is a Senior Research Scientist at Google Brain. Featured References Way Off-Policy Batch Deep Reinforcement Learning of Implicit Human Preferences in Dialog Natasha Jaques, Asma Ghandeharioun, Judy Hanwen Shen, Craig Ferguson, Agata Lapedriza, Noah Jones, Shixiang Gu, Rosalind Picard  Sequence Tutor: Conservative Fine-Tuning of Sequence Generation Models with KL-control Natasha Jaques, Shixiang Gu, Dzmitry Bahdanau, José Miguel Hernández-Lobato, Richard E. Turner, Douglas Eck  PsiPhi-Learning: Reinforcement Learning with Demonstrations using Successor Features and Inverse Temporal Difference Learning Angelos Filos, Clare Lyle, Yarin Gal, Sergey Levine, Natasha Jaques, Gregory Farquhar  Basis for Intentions: Efficient Inverse Reinforcement Learning using Past Experience Marwa Abdulhai, Natasha Jaques, Sergey Levine  Additional References    Fine-Tuning Language Models from Human Preferences, Daniel M. Ziegler et al 2019    Learning to summarize from human feedback, Nisan Stiennon et al 2020    Training language models to follow instructions with human feedback, Long Ouyang et al 2022   

Hear about why OpenAI cites her work in RLHF and dialog models, approaches to rewards in RLHF, ChatGPT, Industry vs Academia, PsiPhi-Learning, AGI and more!  Dr Natasha Jaques is a Senior Research Scientist at Google Brain. Featured References <a href="https://arxiv.org/abs/1907.00456" rel="nofollow" target="_blank">Way Off-Policy Batch Deep Reinforcement Learning of Implicit Human Preferences in Dialog </a>Natasha Jaques, Asma Ghandeharioun, Judy Hanwen Shen, Craig Ferguson, Agata Lapedriza, Noah Jones, Shixiang Gu, Rosalind Picard  <a href="https://arxiv.org/abs/1611.02796" rel="nofollow" target="_blank">Sequence Tutor: Conservative Fine-Tuning of Sequence Generation Models with KL-control </a>Natasha Jaques, Shixiang Gu, Dzmitry Bahdanau, José Miguel Hernández-Lobato, Richard E. Turner, Douglas Eck  <a href="https://arxiv.org/abs/2102.12" rel="nofollow" target="_blank">PsiPhi-Learning: Reinforcement Learning with Demonstrations using Successor Features and Inverse Temporal Difference Learning </a>Angelos Filos, Clare Lyle, Yarin Gal, Sergey Levine, Natasha Jaques, Gregory Farquhar  <a href="https://arxiv.org/abs/2208.04919" rel="nofollow" target="_blank">Basis for Intentions: Efficient Inverse Reinforcement Learning using Past Experience </a>Marwa Abdulhai, Natasha Jaques, Sergey Levine  Additional References  <ul> <li> <a href="https://arxiv.org/abs/1909.08593" rel="nofollow" target="_blank">Fine-Tuning Language Models from Human Preferences</a>, Daniel M. Ziegler et al 2019  </li> <li> <a href="https://arxiv.org/abs/2009.01325" rel="nofollow" target="_blank">Learning to summarize from human feedback</a>, Nisan Stiennon et al 2020  </li> <li> <a href="https://arxiv.org/abs/2203.02155" rel="nofollow" target="_blank">Training language models to follow instructions with human feedback</a>, Long Ouyang et al 2022  </li> </ul>

Natasha Jaques 2

Natasha Jaques is a PhD candidate at MIT working on affective and social intelligence.  She has interned with DeepMind and Google Brain, and was an OpenAI Scholars mentor.  Her paper “Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning” received an honourable mention for best paper at ICML 2019. Featured References Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning Natasha Jaques, Angeliki Lazaridou, Edward Hughes, Caglar Gulcehre, Pedro A. Ortega, DJ Strouse, Joel Z. Leibo, Nando de Freitas Tackling climate change with Machine Learning David Rolnick, Priya L. Donti, Lynn H. Kaack, Kelly Kochanski, Alexandre Lacoste, Kris Sankaran, Andrew Slavin Ross, Nikola Milojevic-Dupont, Natasha Jaques, Anna Waldman-Brown, Alexandra Luccioni, Tegan Maharaj, Evan D. Sherwin, S. Karthik Mukkavilli, Konrad P. Kording, Carla Gomes, Andrew Y. Ng, Demis Hassabis, John C. Platt, Felix Creutzig, Jennifer Chayes, Yoshua Bengio Additional References   MIT Media Lab Flight Offsets,  Caroline Jaffe, Juliana Cherston, Natasha Jaques   Modeling Others using Oneself in Multi-Agent Reinforcement Learning, Roberta Raileanu, Emily Denton, Arthur Szlam, Rob Fergus   Inequity aversion improves cooperation in intertemporal social dilemmas,  Edward Hughes, Joel Z. Leibo, Matthew G. Phillips, Karl Tuyls, Edgar A. Duéñez-Guzmán, Antonio García Castañeda, Iain Dunning, Tina Zhu, Kevin R. McKee, Raphael Koster, Heather Roff, Thore Graepel   Sequential Social Dilemma Games on github, Eugene Vinitsky, Natasha Jaques    AI Alignment newsletter, Rohin Shah   Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutions, Rui Wang, Joel Lehman, Jeff Clune, Kenneth O. Stanley   The social function of intellect, Nicholas Humphrey   Autocurricula and the Emergence of Innovation from Social Interaction: A Manifesto for Multi-Agent Intelligence Research, Joel Z. Leibo, Edward Hughes, Marc Lanctot, Thore Graepel   A Recipe for Training Neural Networks, Andrej Karpathy   Emotionally Adaptive Intelligent Tutoring Systems using POMDPs, Natasha Jaques   Sapiens, Yuval Noah Harari   

Natasha Jaques is a PhD candidate at MIT working on affective and social intelligence.  She has interned with DeepMind and Google Brain, and was an OpenAI Scholars mentor.  Her paper “<a href="https://arxiv.org/abs/1810.08647" rel="nofollow" target="_blank">Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning</a>” received an honourable mention for best paper at ICML 2019. Featured References <a href="https://arxiv.org/abs/1810.08647" rel="nofollow" target="_blank">Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning</a> Natasha Jaques, Angeliki Lazaridou, Edward Hughes, Caglar Gulcehre, Pedro A. Ortega, DJ Strouse, Joel Z. Leibo, Nando de Freitas <a href="https://arxiv.org/abs/1906.05433" rel="nofollow" target="_blank">Tackling climate change with Machine Learning</a> David Rolnick, Priya L. Donti, Lynn H. Kaack, Kelly Kochanski, Alexandre Lacoste, Kris Sankaran, Andrew Slavin Ross, Nikola Milojevic-Dupont, Natasha Jaques, Anna Waldman-Brown, Alexandra Luccioni, Tegan Maharaj, Evan D. Sherwin, S. Karthik Mukkavilli, Konrad P. Kording, Carla Gomes, Andrew Y. Ng, Demis Hassabis, John C. Platt, Felix Creutzig, Jennifer Chayes, Yoshua Bengio  Additional References <ul> <li> <a href="https://offset.media.mit.edu/" rel="nofollow" target="_blank">MIT Media Lab Flight Offsets</a>,  Caroline Jaffe, Juliana Cherston, Natasha Jaques </li> <li> <a href="https://arxiv.org/abs/1802.09640" rel="nofollow" target="_blank">Modeling Others using Oneself in Multi-Agent Reinforcement Learning</a>,  Roberta Raileanu, Emily Denton, Arthur Szlam, Rob Fergus </li> <li> <a href="https://arxiv.org/abs/1803.08884" rel="nofollow" target="_blank">Inequity aversion improves cooperation in intertemporal social dilemmas</a>,   Edward Hughes, Joel Z. Leibo, Matthew G. Phillips, Karl Tuyls, Edgar A. Duéñez-Guzmán, Antonio García Castañeda, Iain Dunning, Tina Zhu, Kevin R. McKee, Raphael Koster, Heather Roff, Thore Graepel </li> <li> <a href="https://github.com/eugenevinitsky/sequential_social_dilemma_games" rel="nofollow" target="_blank">Sequential Social Dilemma Games</a> on github, Eugene Vinitsky, Natasha Jaques  </li> <li> <a href="https://rohinshah.com/alignment-newsletter/" rel="nofollow" target="_blank">AI Alignment newsletter</a>, Rohin Shah </li> <li> <a href="https://arxiv.org/abs/1901.01753" rel="nofollow" target="_blank">Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutions</a>, Rui Wang, Joel Lehman, Jeff Clune, Kenneth O. Stanley </li> <li> <a href="http://cogprints.org/2694/1/SocialFunctionTxt.pdf" rel="nofollow" target="_blank">The social function of intellect</a>, Nicholas Humphrey </li> <li> <a href="https://arxiv.org/abs/1903.00742" rel="nofollow" target="_blank">Autocurricula and the Emergence of Innovation from Social Interaction: A Manifesto for Multi-Agent Intelligence Research</a>, Joel Z. Leibo, Edward Hughes, Marc Lanctot, Thore Graepel </li> <li> <a href="http://karpathy.github.io/2019/04/25/recipe/" rel="nofollow" target="_blank">A Recipe for Training Neural Networks</a>, Andrej Karpathy </li> <li> <a href="https://www.cs.ubc.ca/~jaquesn/POMDPPaper.pdf" rel="nofollow" target="_blank">Emotionally Adaptive Intelligent Tutoring Systems using POMDPs</a>, Natasha Jaques </li> <li> <a href="https://www.ynharari.com/book/sapiens/" rel="nofollow" target="_blank">Sapiens</a>, Yuval Noah Harari  </li> </ul>