When a Correct Reward Is Not Enough: Diagnosing and Guiding PPO in an Analytically Solved Broker-Trader Game
q-fin.TR, cs.AI
Submitted: 2026-10-02
Updated: 2026-10-02
Comments: 8 pages; accepted for publication at ICAIF 2026
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- What Matters In On-Policy Reinforcement Learning? A Large-Scale Empirical Study
- Memory-Induced Supra-Competitive Outcomes Between Deep Reinforcement Learning Agents in Optimal Trade Execution
- Reinforcement Learning for Optimal Execution when Liquidity is Time-Varying
- Bridging Physics-Informed Neural Networks with Reinforcement Learning: Hamilton-Jacobi-Bellman Proximal Policy Optimization (HJBPPO)
- Proximal Policy Optimization Algorithms
- Designing Rewards for Fast Learning
- Broker-Trader Partial Information Nash-Equilibria