Paper: arXiv 2508.16588

Authors: Ziyi Wang, Carmine Ventre, Maria Polukarov

Abstract

Market making is a popular trading strategy, which aims to generate profit from the spread between the quotes posted at either side of the market. It has been shown that training market makers (MMs) with adversarial reinforcement learning allows to overcome the risks due to changing market conditions and to lead to robust performances. Prior work assumes, however, that MMs keep quoting throughout the trading process, but in practice this is not required, even for ``registered’’ MMs (that only need to satisfy quoting ratios defined by the market rules). In this paper, we build on this line of work and enrich the strategy space of the MM by allowing to occasionally not quote or provide single-sided quotes. Towards this end, in addition to the MM agents that provide continuous bid-ask quotes, we have designed two new agents with increasingly richer action spaces. The first has the option to provide bid-ask quotes or refuse to quote. The second has the option to provide bid-ask quotes, refuse to quote, or only provide single-sided ask or bid quotes. We employ a model-driven approach to empirically compare the performance of the continuously quoting MM with the two agents above in various types of adversarial environments. We demonstrate how occasional refusal to provide bid-ask quotes improves returns and/or Sharpe ratios. The quoting ratios of well-trained MMs can basically meet any market requirements, reaching up to 99.9\(%\) in some cases.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced mathematical concepts like stochastic control, limit order book models, and adversarial reinforcement learning, requiring dense formalism. It also presents a model-driven empirical approach with sensitivity analysis on environmental variables, backtesting in simulated adversarial environments, and specific performance metrics like Sharpe ratios.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Enhance Market Making Robustness<br>by Expanding Action Space"] --> B["Methodology:<br>Adversarial Reinforcement Learning<br>with 3 MM Agents"]
  B --> C["Key Data/Inputs:<br>Various Adversarial Market Environments<br>Historical Equities Data"]
  C --> D["Computational Process:<br>Train & Compare 3 Agents:<br>1. Continuous Bid-Ask (Baseline)<br>2. Quote or Refuse<br>3. Quote, Refuse, or Single-Sided"]
  D --> E["Key Findings & Outcomes:<br>• Occasional refusal improves returns & Sharpe ratio<br>• Quoting ratios meet requirements (up to 99.9%)<br>• Enlarged action space yields robust performance"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a flexible market‑making strategy that can optionally not quote or provide single‑sided quotes, extending the action space of market makers.

Source sentence
In this paper, we build on this line of work and enrich the strategy space of the MM by allowing to occasionally not quote or provide single-sided quotes.

Assumes 4

  • Registered market makers must satisfy quoting‑ratio requirements (e.g., 90% of continuous trading). market
    Source sentence
    London Stock exchange requires “registered MM” to maintain a quote “for at least 90% of continuous trading during the mandatory period”
  • When the market maker decides not to quote, the bid/ask offset is set to +∞. model
    Source sentence
    for a trained market maker that decides not to quote, we let the offset be large above a certain threshold at any such time step
  • Order arrivals follow a Poisson process with intensity λ±n = A±n e^{-k±n δ±n}. execution
    Source sentence
    At a given time step, we consider the probability of executing a trade
  • The market environment is adversarially perturbed within bounded ranges for drift, order arrival rates, and volume distribution. other
    Source sentence
    We also examine their performance under different risk coefficients. The results show that training market makers with adversarial reinforcement learning can effectively improve their Sharpe ratio.

Data

Synthetic data only.

Costs

Not modeled A trading or execution result is reported without costs.

Finds 2

  • Allowing market makers to refuse to quote or provide one‑sided quotes improves returns and Sharpe ratios across adversarial settings. performance
    Source sentence
    allowing market makers to provide one-sided quotes and refuse quotes is also an effective means of mitigating market risks.
  • Adversarial reinforcement learning training increases Sharpe ratios compared to baseline models. performance
    Source sentence
    The results show that training market makers with adversarial reinforcement learning can effectively improve their Sharpe ratio.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 15 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive (1 by independent authors)

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.