Beyond Supra-Competitive Outcomes: Collusive Behaviour in Deep Reinforcement Learning for Optimal Execution Games

arXiv:2610.00619 · q-fin.TR, cs.AI · Submitted 2026-09-30 · Read on arXiv

q-fin.TR, cs.AI

Submitted: 2026-09-30

Updated: 2026-09-30

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Deep Reinforcement Learning — 1× in this paper · explained in 3 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes

Sources

• Reinforcement Learning in Queue-Reactive Models: Application to Optimal Execution
• Memory-Induced Supra-Competitive Outcomes Between Deep Reinforcement Learning Agents in Optimal Trade Execution
• Proximal Policy Optimization Algorithms
• AI agents in Algorithmic Electricity Markets: On the Emergence of Tacit Collusion
• Mitigating Retaliatory Algorithmic Collusion in Repeated Games

Related papers

• Mixture-of-Experts for Cryptocurrency Order Execution: Training Stability, Tail Risk, and Failure Modes
• ContestTrade: A Multi-Agent Trading System Based on Internal Contest Mechanism
• Deep reinforcement learning for optimal trading with partial information