TalkRL: The Reinforcement Learning Podcast cover image

NeurIPS 2024 - Posters and Hallways 1

TalkRL: The Reinforcement Learning Podcast

00:00

Intro

This chapter explores a new method for unsupervised skills discovery in hierarchical reinforcement learning, addressing the difficulties of skill selection in specific states without explicit rewards. Through a practical example of driving a car, it demonstrates how a structured skill space can improve performance in various tasks.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app