Mutual Information State Intrinsic Control
iclr.cc · 356 words · saved by 1 readers
N/A
Mutual Information State Intrinsic Control Rui Zhao1,2, Yang Gao3, Pieter Abbeel4, Volker Tresp1,2, Wei Xu5 1. Ludwig Maximilian University of Munich, 2. Siemens AG, 3. Tsinghua University, 4. UC Berkeley, 5. Horizon Robotics 1 Mutual Information State Intrinsic Control International Conference on Learning Representations (ICLR) 2021, Spotlight Motivation and Contribution - Learning by interacting with the environment - Like the way humans learn -…
saved by
related reading
- Multi-task curriculum learning in a complex, visual,hard-exploration domain: Minecraftarxiv.org
- Is Curiosity All You Need? On the Utility of Emergent Behaviours from Curious Explorationarxiv.org
- Reward is not the optimization target — LessWronglesswrong.com
- pdfopenreview.net
- Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Reviewarxiv.org
- [1611.01779] Learning to Act by Predicting the Futurearxiv.org
- 2006.10742arxiv.org
- Complex behavior from intrinsic motivation to occupy future action-state path spacenature.com
- Reward Is Not Enough — LessWronglesswrong.com
- [1705.05363] Curiosity-driven Exploration by Self-supervised Predictionar5iv.labs.arxiv.org
- Ch. 21 - Imitation Learningunderactuated.mit.edu
- What is AIXI?jan.leike.name