Recovering Quantitative Models of Human Information Processing with Differentiable Architecture Search

TL;DR

This paper introduces an automated pipeline combining neural architecture search and automatic differentiation to efficiently recover interpretable models of human information processing from data, advancing cognitive modeling.

Contribution

It presents a novel open-source framework that automates model discovery and fitting, enabling scalable and interpretable cognitive models from large datasets.

Findings

Successfully recovers basic motifs from psychophysics models

Effective in modeling learning and decision-making processes

Identifies limitations and future directions for the framework

Abstract

The integration of behavioral phenomena into mechanistic models of cognitive function is a fundamental staple of cognitive science. Yet, researchers are beginning to accumulate increasing amounts of data without having the temporal or monetary resources to integrate these data into scientific theories. We seek to overcome these limitations by incorporating existing machine learning techniques into an open-source pipeline for the automated construction of quantitative models. This pipeline leverages the use of neural architecture search to automate the discovery of interpretable model architectures, and automatic differentiation to automate the fitting of model parameters to data. We evaluate the utility of these methods based on their ability to recover quantitative models of human information processing from synthetic data. We find that these methods are capable of recovering basic quantitative motifs from models of psychophysics, learning and decision making. We also highlight weaknesses of this framework and discuss future directions for their mitigation.