VWAP Execution with Signature-Enhanced Transformers: A Multi-Asset Learning Approach

In this paper I propose a novel approach to Volume Weighted Average Price (VWAP) execution that addresses two key practical challenges: the need for assetspecific model training and the capture of complex temporal dependencies. Building upon my recent work in dynamic VWAP execution [1], I demonstrate that a single neural network trained across multiple assets can achieve performance comparable to or better than traditional asset-specific models. The proposed architecture combines a transformer-based design inspired by [2] with path signatures for capturing geometric features of price-volume trajectories, as in [3]. The empirical analysis, conducted on hourly cryptocurrency trading data from 80 trading pairs, shows that the globallyfitted model with signature features (GFT-Sig) achieves superior performance in both absolute and quadratic VWAP loss metrics compared to asset-specific approaches. Notably, these improvements persist for out-of-sample assets, demonstrating the model's ability to generalize across different market conditions. The results suggest that combining global parameter sharing with signature-based feature extraction provides a scalable and robust approach to VWAP execution, offering significant practical advantages over traditional asset-specific implementations.