Position: Bayesian Deep Learning is Needed in the Age of Large-Scale AI

TL;DR

This paper advocates for the integration of Bayesian deep learning in large-scale AI to improve uncertainty estimation, active learning, and scientific data analysis, highlighting its potential despite current challenges.

Contribution

It emphasizes the importance of Bayesian deep learning in broad AI applications and discusses future directions for combining it with large foundation models.

Findings

BDL enhances uncertainty quantification in AI models.

It offers advantages for active and continual learning tasks.

Challenges in BDL include computational complexity and scalability.

Abstract

In the current landscape of deep learning research, there is a predominant emphasis on achieving high predictive accuracy in supervised tasks involving large image and language datasets. However, a broader perspective reveals a multitude of overlooked metrics, tasks, and data types, such as uncertainty, active and continual learning, and scientific data, that demand attention. Bayesian deep learning (BDL) constitutes a promising avenue, offering advantages across these diverse settings. This paper posits that BDL can elevate the capabilities of deep learning. It revisits the strengths of BDL, acknowledges existing challenges, and highlights some exciting research avenues aimed at addressing these obstacles. Looking ahead, the discussion focuses on possible ways to combine large-scale foundation models with BDL to unlock their full potential.