Algorithmic Collusion in Dynamic Pricing with Deep Reinforcement Learning

By Shidi Deng, Maximilian Schiffer, Martin Bichler

Rating

1423
Battle Count: 85

Relevance

7/10
Provides insights into algorithmic pricing strategies and potential for tacit collusion, relevant for market making and algorithmic trading

Implementation Complexity

6/10
Requires implementation of multiple RL algorithms and market models, but builds on established frameworks

Reproducibility

4/5
Detailed experimental setup and hyperparameters provided, code implementation available

About this paper

Methodology: Deep Reinforcement Learning. Problem types: Reinforcement Learning, Dynamic Pricing, Market Competition.

The interactive Everscope explorer (charts, battles, favorites) loads below.