Work

Mind the Uncertainty: Risk-Aware and Actively Exploring Model-Based Reinforcement Learning

Year: 2023

Type: preprint

Abstract: 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.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2309.05582

Authors: Vlastelica, Marin, Blaes, Sebastian, Pineri, Cristina, Martius, Georg

Institutions: Max Planck Society, Max Planck Institute for Intelligent Systems, University of Tübingen

Language: English

Topic: Reinforcement Learning in Robotics

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: International Max Planck Research School for Environmental, Cellular and Molecular Microbiology, Bundesministerium für Bildung und Forschung, International Max Planck Research School for Advanced Methods in Process and Systems Engineering

Awards: Verbundprojekt TUEAI: Tübingen AI Center, G4980694453

OpenAlex ID: W4386655688

Cites:

Cited by: 1

Related to: 10