Mind the Uncertainty: Risk-Aware and Actively Exploring Model-Based Reinforcement Learning
Not on the map. Why?
Abstract (arXiv)
We introduce a simple but effective method for managing risk in model-based reinforcement learning with trajectory sampling that involves probabilistic safety constraints and balancing of optimism in the face of epistemic uncertainty and pessimism in the face of aleatoric uncertainty of an ensemble of stochastic neural networks.Various experiments indicate that the separation of uncertainties is essential to performing well with data-driven MPC approaches in uncertain and safety-critical control environments.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 22 of its 43 references that are in the map. First posted on arXiv 11 Sep 2023.
Builds on
22 of 43 references.
References · 22
- 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
- 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
- 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
- 2018Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Modelscited by 774
- 2016Deep Exploration via Bootstrapped DQNcited by 527
- 2018Model-Ensemble Trust-Region Policy Optimizationcited by 219
- 2013Risk-sensitive Reinforcement Learningcited by 130
- 2015Model Predictive Path Integral Control using Covariance Variable Importance Samplingcited by 124
- 2015Gaussian Processes for Data-Efficient Learning in Robotics and Controlcited by 98
- 2020Sample-efficient Cross-Entropy Method for Real-time Planningcited by 95
- 2019Exploring Model-based Planning with Policy Networkscited by 90
- 2020Efficient Model-Based Reinforcement Learning through Optimistic Policy Search and Planningcited by 86
10 more
- 2019Estimating Risk and Uncertainty in Deep Reinforcement Learningcited by 70
- 2019Data Efficient Reinforcement Learning for Legged Robotscited by 63
- 2017Data-Efficient Reinforcement Learning with Probabilistic Model Predictive Controlcited by 53
- 2016Learning and Policy Search in Stochastic Dynamical Systems with Bayesian Neural Networkscited by 49
- 2015One-Shot Learning of Manipulation Skills with Online Dynamics Adaptation and Neural Network Priorscited by 26
- 2019An Open Torque-Controlled Modular Robot Architecture for Legged Locomotion Researchcited by 22
- 2019Perceptual Attention-based Predictive Controlcited by 11
- 2021Safe Distributional Reinforcement Learningcited by 10
- 2020Model-Based Generalization Under Parameter Uncertainty Using Path Integral Controlcited by 7
- 2017Uncertainty Averse Pushing with Model Predictive Path Integral Controlcited by 5
Built on by
Citing papers per year: 2025: 1; 2026: 2.
3 papers.