SC23 Proceedings

The International Conference for High Performance Computing, Networking, Storage, and Analysis

Research Posters Archive

Optimizing Uncertainty Quantification of Vision Transformers in Deep Learning on Novel AI Architectures


Authors: Erik Pautsch (Loyola University, Chicago); John LI (University of California, San Diego); Silvio Rizzi (Argonne National Laboratory (ANL)); George Thiruvathukal (Loyola University, Chicago); and Maria Pantoja (California Polytechnic State University, San Luis Obispo)

Abstract: Deep Learning (DL) methods have shown substantial efficacy in computer vision (CV) and natural language processing (NLP). Despite their proficiency, the inconsistency in input data distributions can compromise prediction reliability. This study mitigates this issue by introducing uncertainty evaluations in DL models, thereby enhancing dependability through a distribution of predictions. Our focus lies on the Vision Transformer (ViT), a DL model that harmonizes both local and global behavior. We conduct extensive experiments on the ImageNet-1K dataset, a vast resource with over a million images across 1,000 categories. ViTs, while competitive, are vulnerable to adversarial attacks, making uncertainty estimation crucial for robust predictions.

Our research advances the field by integrating uncertainty evaluations into ViTs, comparing two significant uncertainty estimation methodologies, and expediting uncertainty computations on high-performance computing (HPC) architectures, such as the Cerebras CS-2, SambaNova DataScale, and the Polaris supercomputer, utilizing the MPI4PY package for efficient distributed training.


Best Poster Finalist (BP): no

Poster: PDF
Poster summary: PDF


Back to Poster Archive Listing