Reinforcement Learning and Robotics

Speaker

Nato is a Research Scientist at HuggingFace. He recently finished his Ph.D. at the University of California, Berkeley, Department of Electrical Engineering and Computer Sciences, advised by Professor Kristofer Pister in the Berkeley Autonomous Microsystems Lab and pseudo-advised by Roberto Calandra at Meta AI Research!

PAST: Ph.D. from Berkeley AI; Cornell ECE `17; Intern at DeepMind, Facebook AI Research, Tesla.

GOAL: To understand and develop safe and societally beneficial autonomous systems.

Abstract

Reinforcement learning is a paradigm in machine learning that uses incentives- or “reinforcement”- to drive learning. The learner is conceptualized as an intelligent agent working within a system of rewards and penalties in order to solve a novel problem. The agent is designed to maximize rewards while pursuing a solution by trial-and-error. Programming a system to respond to the complex and unpredictable “real world” is one of the principal challenges in robotics engineering. One field which is finding new applications for reinforcement learning is the study of MEMS devices- robots or other electronic devices built at the micrometer scale. The use of reinforcement learning in microscopic devices poses a challenging engineering problem due to constraints with power usage and computational power.

Topics

0:00 Intro

1:09 Talk

32:57 Tips for getting internships at companies like DeepMind, Facebook, Tesla

34:23 Objective mismatch in reinforcement learning

37:01 Equal Access to Application Assistance Program at Berkeley (MS/Ph.D.) 38:35 Job search as a Ph.D. in ML

42:48 Why is impossible to get into some masters