Learning a functional control for high-frequency finance

TL;DR

This paper introduces a deep neural network approach to generate and explain optimal trading controls in high-frequency finance, addressing challenges like market impact, data scarcity, and regulatory transparency.

Contribution

It presents a novel neural network framework that learns the mapping from trader preferences to controls, incorporating transfer learning and explainability techniques for practical financial applications.

Findings

Neural network successfully learns the mapping between risk preferences and trading controls.

Transfer learning improves training efficiency with simulated trajectories.

Projected controls are transparent and closely match original controls, aiding regulatory acceptance.

Abstract

We use a deep neural network to generate controllers for optimal trading on high frequency data. For the first time, a neural network learns the mapping between the preferences of the trader, i.e. risk aversion parameters, and the optimal controls. An important challenge in learning this mapping is that in intraday trading, trader's actions influence price dynamics in closed loop via the market impact. The exploration--exploitation tradeoff generated by the efficient execution is addressed by tuning the trader's preferences to ensure long enough trajectories are produced during the learning phase. The issue of scarcity of financial data is solved by transfer learning: the neural network is first trained on trajectories generated thanks to a Monte-Carlo scheme, leading to a good initialization before training on historical trajectories. Moreover, to answer to genuine requests of financial regulators on the explainability of machine learning generated controls, we project the obtained "blackbox controls" on the space usually spanned by the closed-form solution of the stylized optimal trading problem, leading to a transparent structure. For more realistic loss functions that have no closed-form solution, we show that the average distance between the generated controls and their explainable version remains small. This opens the door to the acceptance of ML-generated controls by financial regulators.