Natasha Jaques
Assistant Professor at Google
Seattle, Washington, United States
Summary
👤
Senior🎓
Top SchoolNatasha Jaques is an Assistant Professor and Senior Research Scientist with 12 years of experience at the intersection of deep learning, reinforcement learning, and human-AI interaction, currently splitting her time between the University of Washington and Google. Her research focuses on social reinforcement learning—using multi-agent techniques to improve single-agent generalization—and she has a strong publication record documented on Google Scholar. Natasha combines academic rigor from a PhD at MIT and prior roles at DeepMind and Google Brain with hands-on engineering, exemplified by contributions to open-source RL environments (e.g., implementing RGB map rendering and video recording for sequential social dilemmas). Based in Seattle, she brings a unique blend of psychology and computer science training, pairing computational insight with a behavioral perspective on AI systems.
12 years of coding experience
PhD, PhD at Massachusetts Institute of TechnologyMaster of Science (M.Sc.), Computer Science, Master of Science (M.Sc.), Computer Science at The University of British ColumbiaBachelor of Arts (B.A.), Psychology, Bachelor of Arts (B.A.), Psychology at University of Regina
Stackoverflow
1reputation
0reached
1answer
0questions
Github Skills (10)
video-production10
video-editing10
opencv10
video-encoder10
python10
video-conversion10
maps9
ruby6
ruby-on-rails6
actionmailer6
Programming languages (3)
Github contributions (5)
eugenevinitsky/sequential_social_dilemma_games
Dec 2018 - Apr 2019
Repo for reproduction of sequential social dilemmas
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:102 commits, 83 pushes, 2 branches in 3 months
Contributions summary:Natasha primarily focused on implementing environment rendering and video rendering features. They modified the environment's map rendering to use RGB values and added video recording functionality using OpenCV, saving images to disk and assembling them into videos. Furthermore, the user incorporated the ability to override the video rendering path from the command line. The user also performed bug fixes relating to the video generation and general code cleanup.
sequentialreproductionmachine-learningsocial-dilemmas
mitmedialab/PersonalizedMultitaskLearning
Jun 2017 - Aug 2019
Code for performing 3 multitask machine learning methods: deep neural networks, Multitask Multi-kernel Learning (MTMKL), and a hierarchical Bayesian model (HBLR).
Contributions:24 commits, 18 pushes, 1 branch in 2 years 2 months
kerneldeep-learninghierarchicalmachine-learningneural-networks
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.