Challenges of real-world reinforcement learning: definitions, benchmarks and analysis

DOI10.1007/s10994-021-05961-4OpenAlexW3121342653WikidataQ114224937 ScholiaQ114224937MaRDI QIDQ2071388

Nir Levine, Sven Gowal, Todd Hester, Cosmin Paduraru, Daniel J. Mankowitz, Gabriel Dulac-Arnold, Jerry Li

Publication date: 28 January 2022

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10994-021-05961-4

zbMATH Keywords

reinforcement learningapplied reinforcement learningreal-world

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Related Items (6)

Planning for potential: efficient safe reinforcement learning ⋮ Learning reward machines: a study in partially observable reinforcement learning ⋮ Reward (Mis)design for autonomous driving ⋮ Data-driven passivity-based control of underactuated mechanical systems via interconnection and damping assignment ⋮ Safety-constrained reinforcement learning with a distributional safety critic ⋮ Online Bootstrap Inference For Policy Evaluation In Reinforcement Learning

Uses Software

• POMDP
• TEXPLORE
• Horizon
• Safety Gym
• QT-Opt
• IMPALA
• MOPO

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• \(T_{\mathcal{P}}\)-compilation for inference in probabilistic logic programs
• Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning
• A Survey of Multi-Objective Sequential Decision-Making
• Multiple Model-Based Reinforcement Learning
• From Skills to Symbols: Learning Symbolic Representations for Abstract High-Level Planning
• 10.1162/1532443041827907
• Robust Dynamic Programming

This page was built for publication: Challenges of real-world reinforcement learning: definitions, benchmarks and analysis