Newest papers

Cooperative Multi-Agent Reinforcement Learning for Market Making on Prediction-Market Order Books

Skander Lejmi · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26789 · cited by 0 works (OpenAlex) · open-access PDF

Vietnamese

Chúng tôi nghiên cứu các nhà sản xuất thị trường được đào tạo với sự tăng cường đa đại lý học tập về các thị trường dự đoán, đặc biệt là Polymarket, mà các cuốn sách trật tự mỏng đặt ra những thách thức riêng biệt. Ba thách thức hỗ trợ nghiên cứu này: lựa chọn bất lợi có thể gây ra một chính sách được đào tạo trên dữ liệu thực để ngăn chặn báo giá; hậu quả của một báo giá có thể đến nhiều bước sau khi hành động; và dữ liệu dự đoán-thị trường L2 không xác định các đơn đặt hàng riêng lẻ cần thiết cho mô phỏng. Chúng tôi giải quyết những thách thức tích cực này bằng cách tái cấu trúc dữ liệu LOBSTER ở cấp lệnh. Ba thách thức hỗ trợ nghiên cứu này: lựa chọn bất lợi có thể gây ra một chính sách được đào tạo trên dữ liệu thực để ngăn chặn báo giá; chúng tôi làm

Machine translation (Meta M2M100); the English original follows.

Abstract

A market maker provides liquidity by quoting prices to buy and sell an asset, earning the spread while managing inventory risk. We study market makers trained with multi-agent reinforcement learning on prediction markets, specifically Polymarket, whose thin order books pose distinct challenges. Three challenges underpin this research: adverse selection may cause a policy trained on real data to stop quoting; the consequences of a quote can arrive many steps after the action; and prediction-market L2 data does not identify individual orders needed for simulation. We address these challenges by reconstructing order-level LOBSTER-format data from Polymarket and training agents in JaxMARL-HFT on 417 real markets. We enrich the baseline market maker's observation space with the book and recent flow, redesign its actions and inventory controls, and propose a cooperative multi-agent decomposition into a quoter and a risk manager trained using recurrent Independent Proximal Policy Optimisation (IPPO) with RNN policies. We investigate reward shaping, particularly reward blending and post-fill markout, to mitigate credit-assignment difficulties between the agents. Our redesigned single-agent market maker quotes more frequently and incurs smaller held-out losses than the retrained JaxMARL-HFT baseline, although the contributions of jointly changed observations, actions and risk constraints are not separately identified. Decomposition alone performs poorly; careful reward design substantially improves the desk relative to the unshaped split and yields our best result. Models are evaluated on realised end-of-episode portfolio value on held-out markets. None establishes positive average portfolio value in periodic held-out evaluation, and each configuration was trained with one seed. The environment, data and protocol lay a foundation for makers that quote many linked markets at once.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Reinforcement learning, Portfolio, Desk, Computer science, Order (exchange), Market maker

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26789, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange