Catastrophic-risk-aware reinforcement learning with extreme-value-theory-based policy gradients

Parisa Davar, Frédéric Godin, Jose Garrido
6/21/2024
cs.LGq-fin.RM

No official code found. Best implementation: parisadavar/EVT-policy-gradient-RL (2 stars, last commit 2024-06-27).

Abstract

This paper tackles the problem of mitigating catastrophic risk (which is risk with very low frequency but very high severity) in the context of a sequential decision making process. This problem is particularly challenging due to the scarcity of observations in the far tail of the distribution of cumulative costs (negative rewards). A policy gradient algorithm is developed, that we call POTPG. It is based on approximations of the tail risk derived from extreme value theory. Numerical experiments highlight the out-performance of our method over common benchmarks, relying on the empirical distribution. An application to financial risk management, more precisely to the dynamic hedging of a financial option, is presented.

View on arXivView PDF

Code Implementations(1)

parisadavar/EVT-policy-gradient-RLCommunity100%
20Jupyter NotebookMay 16, 20242 years ago

Similar Papers

Policy Gradient Methods for Risk-Sensitive Distributional Reinforcement Learning with Provable Convergence

84% match2024

Robust Risk-Aware Option Hedging

84% match2023

On the Global Convergence of Risk-Averse Policy Gradient Methods with Expected Conditional Risk Measures

84% match2023

Risk-Averse Reinforcement Learning via Dynamic Time-Consistent Risk Measures

83% match2023

EX-DRL: Hedging Against Heavy Losses with EXtreme Distributional Reinforcement Learning

83% match1 repo52024

A Reductions Approach to Risk-Sensitive Reinforcement Learning with Optimized Certainty Equivalents

83% match2024

Cite this paper

@article{davar2024catastrophicriskaware,
  title  = {Catastrophic-risk-aware reinforcement learning with extreme-value-theory-based policy gradients},
  author = {Parisa Davar and Frédéric Godin and Jose Garrido},
  year   = {2024},
  eprint = {2406.15612},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.15612v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.15612.svg)](https://codeofpaper.com/paper/2406.15612)

Discussion