ProbML is a venue for researchers to gather and exchange ideas on the theory and practice of probabilistic machine learning and AI, especially (but not limited to) Bayesian techniques.
ProbML will be held on July 5, 2026 in Seoul, co-located with ICML 2026. This year’s symposium features an expanded scope with two primary areas of focus:
Area 1: Probabilistic and/or Bayesian ML Methods
We encourage submissions that advance the foundations of probabilistic machine learning, probabilistic and (approximate) Bayesian inference, Bayesian statistics, and decision-making under uncertainty. We also welcome work exploring connections between these fields and adjacent areas such as deep learning, natural language processing, active learning, reinforcement learning, compression, AI safety, scientific computing, causal inference, foundation models, and lifelong learning.
Area 2 (New!): Applications of Probabilistic and/or Bayesian Methods with a focus on Healthcare and Climate Change
This new focus area aims to foster stronger communication between methodology researchers and application-focused researchers. We encourage submissions that propose thoughtful, rigorous uses of probabilistic/Bayesian methods with clear empirical and scientific evidence for why the chosen methodology is well-aligned with the downstream task, particularly in data-sparse regimes, high-stakes decision making, or settings with strict interpretability, safety, or governance requirements.
We are delighted to offer three submission tracks:
- a Proceedings Track, for full research papers of up to 9 content pages (archival),
- a Workshop Track, for extended abstracts of 3-5 content pages (non-archival), and
- a Fast Track, for papers recently accepted at major ML venues.
For the full call for papers and submission instructions, click here.