analiza aiqbench
This theoretical paper studies how learning trading agents, modeled by their reward functions, interact with constant-product automated market makers. Its main finding is that herd-driven and arbitrage-driven agent populations can look identical on standard risk and volume metrics yet differ sharply in how long stress episodes last, so serial statistics are needed to diagnose the feedback. For AI readers, the useful takeaway is that multi-agent learning can create fragile, all-or-nothing participation dynamics, and that pricing structure (maker-taker fees) may matter more than uniform fees for stability.
Learning AI Agents, Market Makers, and Stress in Automated Markets
arXiv:2611.01799v1 Announce Type: cross Abstract: Automated markets increasingly route order flow from learning trading agents through automated market makers (AMMs). We model trader archetypes as AI agents defined by their reward functions and analyze the feedback they create on a constant-product AMM. Herd-dominated and arbitrage-dominated markets of equal size have identical return distributions and turnover, so no rule based on volatility, value-at-risk, stress frequency, or volume can tell them apart; yet their stress episodes differ in length by a factor unbounded in the feedback-to-noise ratio $\beta/s$. Serial statistics identify the feedback sign from $\Theta((s/\beta)^2\log(1/\delta))$ observations, and this rate is tight. When agents trade together, profit-seeking agents ride herds rather than lean against them, leading to all-or-nothing participation. Spread-earning market makers restore stability but make liquidity fragile, with discontinuous collapse and hysteresis; maker--taker pricing, not a uniform fee, is the effective instrument.
Abstract:Automated markets increasingly route order flow from learning trading agents through automated market makers (AMMs). We model trader archetypes as AI agents defined by their reward functions and analyze the feedback they create on a constant-product AMM. Herd-dominated and arbitrage-dominated markets of equal size have identical return distributions and turnover, so no rule based on volatility, value-at-risk, stress frequency, or volume can tell them apart; yet their stress episodes differ in length by a factor unbounded in the feedback-to-noise ratio $\beta/s$. Serial statistics identify the feedback sign from $\Theta((s/\beta)^2\log(1/\delta))$ observations, and this rate is tight. When agents trade together, profit-seeking agents ride herds rather than lean against them, leading to all-or-nothing participation. Spread-earning market makers restore stability but make liquidity fragile, with discontinuous collapse and hysteresis; maker--taker pricing, not a uniform fee, is the effective instrument.
Econometrics (econ.EM); Machine Learning (cs.LG); Theoretical Economics (econ.TH); Trading and Market Microstructure (q-fin.TR)
arXiv:2611.01799 [econ.EM]
(or arXiv:2611.01799v1 [econ.EM] for this version)
https://doi.org/10.48550/arXiv.2611.01799 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 240 cuvinte
