The Practicality of Normalizing Flow Test-Time Training in Bayesian Inference for Agent-Based Models

TL;DR

This paper explores the use of test-time training with normalizing flows to improve Bayesian inference in agent-based models, demonstrating real-time parameter adjustment under distribution shifts.

Contribution

It introduces practical TTT strategies for normalizing flows in ABMs and evaluates their effectiveness in real-time inference scenarios.

Findings

TTT schemes significantly improve inference accuracy

Real-time adjustment of flow-based inference is feasible

TTT enhances robustness against distribution shifts

Abstract

Agent-Based Models (ABMs) are gaining great popularity in economics and social science because of their strong flexibility to describe the realistic and heterogeneous decisions and interaction rules between individual agents. In this work, we investigate for the first time the practicality of test-time training (TTT) of deep models such as normalizing flows, in the parameters posterior estimations of ABMs. We propose several practical TTT strategies for fine-tuning the normalizing flow against distribution shifts. Our numerical study demonstrates that TTT schemes are remarkably effective, enabling real-time adjustment of flow-based inference for ABM parameters.