Zero-shot adaptation to order book dynamics
This work addresses the need for adaptive market-making strategies that can handle non-stationary order book dynamics, offering a computationally efficient solution for practitioners in quantitative finance.
The paper proposes an adaptive market-making architecture that retains the analytical structure of the Avellaneda-Stoikov framework while enabling zero-shot adaptation to changing market regimes and trading objectives via a successor measure-style mechanism. The approach separates market dynamics from trading objectives, allowing real-time adjustment of bid-ask quotes without retraining.
We describe an adaptive market-making architecture that preserves the analytical structure of the Avellaneda--Stoikov framework while introducing a successor measure-style adaptation mechanism. In our paper we keep Avellaneda--Stoikov fast Hamilton--Jacobi--Bellman structure and make it adaptive to changing market regimes and trading objectives. The central idea is to separate market dynamics from the trading objective. The market state determines a low-dimensional set of Avellaneda--Stoikov parameters, while recent realized rewards determine a low-dimensional objective vector. The HJB forward map then converts this objective into optimal bid and ask quotes through a scalarization of future reward features.