all AI news
Exploration is Harder than Prediction: Cryptographically Separating Reinforcement Learning from Supervised Learning
April 8, 2024, 4:42 a.m. | Noah Golowich, Ankur Moitra, Dhruv Rohatgi
cs.LG updates on arXiv.org arxiv.org
Abstract: Supervised learning is often computationally easy in practice. But to what extent does this mean that other modes of learning, such as reinforcement learning (RL), ought to be computationally easy by extension? In this work we show the first cryptographic separation between RL and supervised learning, by exhibiting a class of block MDPs and associated decoding functions where reward-free exploration is provably computationally harder than the associated regression problem. We also show that there is …
abstract arxiv cs.cc cs.cr cs.ds cs.lg easy exploration extension mean practice prediction reinforcement reinforcement learning show supervised learning type work
More from arxiv.org / cs.LG updates on arXiv.org
Jobs in AI, ML, Big Data
AI Research Scientist
@ Vara | Berlin, Germany and Remote
Data Architect
@ University of Texas at Austin | Austin, TX
Data ETL Engineer
@ University of Texas at Austin | Austin, TX
Lead GNSS Data Scientist
@ Lurra Systems | Melbourne
Senior Machine Learning Engineer (MLOps)
@ Promaton | Remote, Europe
Robotics Technician - 3rd Shift
@ GXO Logistics | Perris, CA, US, 92571