Risk Aware Reinforcement Learning

Risk-aware reinforcement learning (RARL) aims to develop agents that make optimal decisions while explicitly considering and mitigating potential risks, going beyond simply maximizing expected rewards. Current research focuses on incorporating various risk measures, such as quantile regression, dynamic distortion risk measures, and extreme value theory, into reinforcement learning algorithms like actor-critic methods, often employing neural networks for function approximation. This field is significant because it enables the deployment of more robust and reliable AI agents in high-stakes applications, such as financial trading, robotics, and resource management, where unforeseen events can have severe consequences.

Papers

November 5, 2024

When to Localize? A Risk-Constrained Reinforcement Learning Approach
Chak Lam Shek, Kasra Torshizi, Troi Williams, Pratap Tokekar
Reinforcement Learning Observable Markov Decision Process Underwater Robot Risk Aware Reinforcement Learning Active Localization

September 16, 2024

June 21, 2024

Catastrophic-risk-aware reinforcement learning with extreme-value-theory-based policy gradients
Parisa Davar, Frédéric Godin, Jose Garrido
Policy Gradient Sequential Decision Extreme Value Theory Policy Gradient Algorithm Tail Risk Risk Aware Reinforcement Learning Dynamic Hedging

May 29, 2024

Spectral-Risk Safe Reinforcement Learning with Convergence Guarantees
Dohyeong Kim, Taehyun Cho, Seungyub Han, Hojun Chung, Kyungjae Lee, Songhwai Oh
Reinforcement Learning Bilevel Optimization Convergence Guarantee Risk Aware Reinforcement Learning

November 5, 2023

Multi-objective Reinforcement Learning with Nonlinear Preferences: Provable Approximation for Maximizing Expected Scalarized Return
Nianli Peng, Muhang Tian, Brandon Fain
Multi Objective Reinforcement Learning Provable Guarantee Risk Aware Reinforcement Learning Multi Objective Markov Decision Process Reward Aware

October 30, 2023

On the Theory of Risk-Aware Agents: Bridging Actor-Critic and Economics
Michal Nauman, Marek Cygan
Theoretical Understanding Utility Maximization Prospect Theory Risk Aware Reinforcement Learning Risk Aware Policy

September 12, 2023

Risk-Aware Reinforcement Learning through Optimal Transport Theory
Ali Baheri
Reinforcement Learning Optimal Transport Risk Aware Reward Maximization Reinforcement Learning Objective Risk Aware Reinforcement Learning

August 24, 2023

Extreme Risk Mitigation in Reinforcement Learning using Extreme Value Theory
Karthik Somayaji NS, Yu Wang, Malachi Schram, Jan Drgona, Mahantesh Halappanavar, Frank Liu, Peng Li
Reinforcement Learning Risk Sensitive Reinforcement Learning Risk Management Extreme Value Theory State Action Distribution Risk Aware Reinforcement Learning State Action Value

April 30, 2023

A Transfer Learning Approach to Minimize Reinforcement Learning Risks in Energy Optimization for Smart Buildings
Mikhail Genkin, J. J. McArthur
Deep Reinforcement Learning Reinforcement Learning Agent Energy Efficiency Transfer Learning Approach Smart Building Risk Aware Reinforcement Learning

April 1, 2023

Mastering Pair Trading with Risk-Aware Recurrent Reinforcement Learning
Weiguang Han, Jimin Huang, Qianqian Xie, Boyi Zhang, Yanzhao Lai, Min Peng
Reinforcement Learning Risk Aware Reinforcement Learning Pair Trading

March 27, 2023

Robust Risk-Aware Option Hedging
David Wu, Sebastian Jaimungal
Hedging Strategy Robust Strategy Risk Aware Reinforcement Learning Dynamic Hedging Path Dependent Option

August 23, 2022

An intelligent algorithmic trading based on a risk-return reinforcement learning algorithm
Boyi Jin
Deep Reinforcement Learning Risk Aware Reinforcement Learning Portfolio Optimization Model Intelligent Trading Prediction Sorted Portfolio

December 15, 2021

DeepScalper: A Risk-Aware Reinforcement Learning Framework to Capture Fleeting Intraday Trading Opportunities
Shuo Sun, Wanqi Xue, Rundong Wang, Xu He, Junlei Zhu, Jian Li, Bo An
Reinforcement Learning Risk Aware Reinforcement Learning Intraday Market