Podcasts Technology The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Safer Exploration in Deep Reinforcement Learning using Action Priors with Sicelukwanda Zwane - TWiML Talk #235

Safer Exploration in Deep Reinforcement Learning using Action Priors with Sicelukwanda Zwane - TWiML Talk #235

Published: March 1, 2019, 5 p.m.

b'Today we conclude our Black in AI series with Sicelukwanda Zwane, a masters student at the University of Witwatersrand and graduate research assistant at the CSIR, who presented on \\u201cSafer Exploration in Deep Reinforcement Learning using Action Priors\\u201d at the workshop. In our conversation, we discuss what \\u201csafer exploration\\u201d means in this sense, the difference between this work and other techniques like imitation learning, and how this fits in with the goal of \\u201clifelong learning.\\u201d'