NCLGAug 17, 2022

"Task-relevant autoencoding" enhances machine learning for human neuroscience

arXiv:2208.08478v21 citationsh-index: 96
Originality Incremental advance
AI Analysis

This addresses the challenge of analyzing high-dimensional, small-sample neuroimaging data for researchers in human neuroscience, though it appears incremental as it builds on existing autoencoder methods.

The authors tackled the problem of overfitting in machine learning models for human neuroscience by developing TRACE, a task-relevant autoencoder, which outperformed standard methods with up to 12% increased classification accuracy and up to 56% improvement in extracting behaviorally-relevant representations from fMRI data.

In human neuroscience, machine learning can help reveal lower-dimensional neural representations relevant to subjects' behavior. However, state-of-the-art models typically require large datasets to train, so are prone to overfitting on human neuroimaging data that often possess few samples but many input dimensions. Here, we capitalized on the fact that the features we seek in human neuroscience are precisely those relevant to subjects' behavior. We thus developed a Task-Relevant Autoencoder via Classifier Enhancement (TRACE), and tested its ability to extract behaviorally-relevant, separable representations compared to a standard autoencoder, a variational autoencoder, and principal component analysis for two severely truncated machine learning datasets. We then evaluated all models on fMRI data from 59 subjects who observed animals and objects. TRACE outperformed all models nearly unilaterally, showing up to 12% increased classification accuracy and up to 56% improvement in discovering "cleaner", task-relevant representations. These results showcase TRACE's potential for a wide variety of data related to human behavior.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes