Entropy Regularized Reinforcement Learning for Zero-Sum Stochastic Differential Games in a Regime-Switching Jump-Diffusion Process

• URL: http://arxiv.org/abs/2606.28669v1
• Date: Sat, 27 Jun 2026 01:03:53 GMT
• Title: Entropy Regularized Reinforcement Learning for Zero-Sum Stochastic Differential Games in a Regime-Switching Jump-Diffusion Process
• Abstract summary: This paper introduces a distributional control approach that characterizes optimal strategies as probability distributions over actions.<n>It forms a reinforcement learning framework for entropy-regularized differential games.<n> numerical examples illustrate the effect of the temperature parameter and regime transitions on optimal policies and values.
• Score: 3.395785516451296
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: To address parameter misspecification and sudden structural environmental changes in conventional stochastic differential game (SDG) frameworks, this paper introduces a distributional control approach that characterizes optimal strategies as probability distributions over actions, conditioned on the continuous state, the discrete regime state, and parameters. This forms a reinforcement learning framework for entropy-regularized zero-sum stochastic differential games (ERRL-ZSSDGs) in a regime-switching jump-diffusion process. Using the dynamic programming principle (DPP), we derive the associated coupled systems of Hamilton-Jacobi-Bellman-Isaacs (HJBI) equations, from which equilibrium strategies are expressed via gradients of the value function. For linear-quadratic problems, semi-analytical solutions for both value function and equilibrium strategies are obtained by solving a system of coupled ordinary differential equations (ODEs). In more general settings, an Actor-Critic policy improvement algorithm is developed to approximate the value functions and equilibrium policies across different regimes. The method is applied to an investment game, and numerical examples illustrate the effect of the temperature parameter and regime transitions on optimal policies and values.

Related papers

• Entropy-Regularized Reinforcement Learning for Linear-Quadratic Stackelberg Differential Games in Regime-Switching Diffusion Models [5.600703864320554]
  This paper proposes an entropy-regularized reinforcement learning approach for Stack differential games.<n>The key innovation lies in deriving weakly-coupled HJBI equations with entropy regularization.<n>Neural networks are integrated to approximate regime-dependent value functions and solve high-dimensional partial differential equations.
  arXiv  Detail & Related papers  (2026-06-27T01:17:22Z)
• From Noise to Control: Parameterized Diffusion Policies [54.941343992110426]
  We propose a framework for learning diffusion policies conditioned on continuous parameters embedded in a learned behavior manifold.<n>Our approach enables smooth distances between known strategies and efficient adaptation to novel constraints without updating policy weights.
  arXiv  Detail & Related papers  (2026-05-29T20:21:50Z)
• Continuous-time reinforcement learning for optimal switching over multiple regimes [5.045537244224327]
  This paper studies the continuous-time reinforcement learning (RL) for optimal switching problems across multiple regimes.<n>We establish the well-posedness of the associated system of Hamilton-Jacobi-Bellman equations and provide a characterization of the optimal policy.<n>A reinforcement learning algorithm is devised and implemented by invoking the policy evaluation based on the martingale characterization.
  arXiv  Detail & Related papers  (2025-12-04T11:48:07Z)
• On the Noisy Road to Open Quantum Dynamics: The Place of Stochastic Hamiltonians [0.0]
  evolution underpins several approaches to the dynamics of open quantum systems.<n>In a formulation, the open-system problem is reduced from a coupled system-environment to an effective system-only description.
  arXiv  Detail & Related papers  (2025-10-11T09:40:29Z)
• Stabilization of jump-diffusion stochastic differential equations by hysteresis switching [1.864621482724548]
  We address the stabilization of both classical and quantum systems modeled by jump-diffusion differential equations.<n>Our approach employs local Lyapunov-like conditions and state-dependent switching to achieve global or exponential stability.
  arXiv  Detail & Related papers  (2025-07-21T02:24:00Z)
• Solving wave equation problems on D-Wave quantum annealers [44.99833362998488]
  We solve the one-dimensional Helmholtz equation in several scenarios using the quantum annealer provided by the D-Wave systems within a pseudospectral scheme.<n>We assess the performance of different strategies of encoding based on algebraic arguments and the adiabatic condition.
  arXiv  Detail & Related papers  (2025-07-18T08:06:43Z)
• On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning [59.11784194183928]
  Policy gradient algorithms have been successfully applied to enhance the reasoning capabilities of large language models (LLMs)<n>Regularized Policy Gradient (RPG) view shows that the widely-used $k_3$ penalty is exactly the unnormalized KL.<n>RPG-REINFORCE with RPG-Style Clip improves accuracy by up to $+6$ absolute percentage points over DAPO.
  arXiv  Detail & Related papers  (2025-05-23T06:01:21Z)
• Learning Controlled Stochastic Differential Equations [61.82896036131116]
  This work proposes a novel method for estimating both drift and diffusion coefficients of continuous, multidimensional, nonlinear controlled differential equations with non-uniform diffusion. We provide strong theoretical guarantees, including finite-sample bounds for (L2), (Linfty), and risk metrics, with learning rates adaptive to coefficients' regularity. Our method is available as an open-source Python library.
  arXiv  Detail & Related papers  (2024-11-04T11:09:58Z)
• Stochastic Q-learning for Large Discrete Action Spaces [79.1700188160944]
  In complex environments with discrete action spaces, effective decision-making is critical in reinforcement learning (RL) We present value-based RL approaches which, as opposed to optimizing over the entire set of $n$ actions, only consider a variable set of actions, possibly as small as $mathcalO(log(n)$)$. The presented value-based RL methods include, among others, Q-learning, StochDQN, StochDDQN, all of which integrate this approach for both value-function updates and action selection.
  arXiv  Detail & Related papers  (2024-05-16T17:58:44Z)
• Ensemble Kalman Filtering Meets Gaussian Process SSM for Non-Mean-Field and Online Inference [47.460898983429374]
  We introduce an ensemble Kalman filter (EnKF) into the non-mean-field (NMF) variational inference framework to approximate the posterior distribution of the latent states. This novel marriage between EnKF and GPSSM not only eliminates the need for extensive parameterization in learning variational distributions, but also enables an interpretable, closed-form approximation of the evidence lower bound (ELBO) We demonstrate that the resulting EnKF-aided online algorithm embodies a principled objective function by ensuring data-fitting accuracy while incorporating model regularizations to mitigate overfitting.
  arXiv  Detail & Related papers  (2023-12-10T15:22:30Z)
• Machine learning in and out of equilibrium [58.88325379746631]
  Our study uses a Fokker-Planck approach, adapted from statistical physics, to explore these parallels. We focus in particular on the stationary state of the system in the long-time limit, which in conventional SGD is out of equilibrium. We propose a new variation of Langevin dynamics (SGLD) that harnesses without replacement minibatching.
  arXiv  Detail & Related papers  (2023-06-06T09:12:49Z)
• Structure-Preserving Learning Using Gaussian Processes and Variational Integrators [62.31425348954686]
  We propose the combination of a variational integrator for the nominal dynamics of a mechanical system and learning residual dynamics with Gaussian process regression. We extend our approach to systems with known kinematic constraints and provide formal bounds on the prediction uncertainty.
  arXiv  Detail & Related papers  (2021-12-10T11:09:29Z)
• Implicit Distributional Reinforcement Learning [61.166030238490634]
  implicit distributional actor-critic (IDAC) built on two deep generator networks (DGNs) Semi-implicit actor (SIA) powered by a flexible policy distribution. We observe IDAC outperforms state-of-the-art algorithms on representative OpenAI Gym environments.
  arXiv  Detail & Related papers  (2020-07-13T02:52:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.