js05212/BayesianDeepLearning-Survey

What it solves

Bayesian deep learning is a fast-moving research area, and it is easy to miss important work. This repository acts as a structured, continuously updated reference that collects papers across many application fields, so you can quickly see what has been done and where. It builds on a formal survey published in ACM Computing Surveys (2020) and extends it with newer research.

How it works

The project is a GitHub repository whose README is a categorized list of papers. Each entry gives the paper title, authors, venue, and a link to the PDF (and sometimes code or slides). The categories cover major application areas: recommender systems, domain adaptation and meta-learning, healthcare, natural language processing, computer vision, control and planning, graphs, topic modeling, speech, forecasting, federated learning, continual learning, AI for science, and miscellaneous. Some papers appear in multiple categories and are cross-referenced. The list is updated over time, as the title says, "An Updating Survey".

Who it's for

This is aimed at researchers, graduate students, and practitioners in machine learning and deep learning who want a comprehensive, up-to-date map of Bayesian deep learning literature. It is especially useful for anyone entering the field or looking for related work in a specific application area.

Highlights

• Based on a peer-reviewed survey published in ACM Computing Surveys, 2020.
• Covers a wide range of application domains, from healthcare to computer vision to natural language processing.
• Each entry includes a direct PDF link, making it easy to access the full text.
• Cross-references between related sections help you follow work that spans multiple topics.
• The repository is explicitly described as "constantly updated and extended," so it tracks new developments beyond the original survey.