EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

2493. Reinforcement Learning Methods in Risk-Sensitive Investment Management

Invited abstract in session MC-57: Dynamic portfolio selection: stochastic optimization, filtering, and learning techniques, stream Modern Decision Making in Finance and Insurance.

Monday, 12:30-14:00
Room: S06 (building: 101)

Authors (first author is the speaker)

1. Wolfgang Runggaldier
Mathematics, University of Padova
2. Sebastien Lleo
Finance Department, NEOMA Business School

Abstract

We investigate the benefits of relating reinforcement learning (RL) with risk-sensitive control. Our starting point is the duality between free energy and relative entropy, see e.g. Dai Pra et al. (1996). It establishes an equivalence between risk-sensitive control and standard stochastic control problems with an entropy regularization term.
This approach has two major advantages:
i) it does not require a preliminary change of measure à la Kuroda and Nagai (2002);
ii) it is naturally consistent with the use of a regularization/penalization term in the literature that connects reinforcement learning with stochastic control, e.g. Wang et al (2020). In this sense it also allows for a risk-sensitive interpretation of the entropy regularization in RL.

We furthermore show how this connects to the existing literature on risk-sensitive investment management (Kuroda and Nagai, 2002, Davis and Lleo, 2008, 2020, 2021), whereby cases with unknown parameters or with partial observation showcase the advantages of reinforcement learning.

Keywords

Status: accepted

Back to the list of papers