v289: Proceedings of AABI 2025

2026-06-04 · Source: Proceedings of Machine Learning Research · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, quick

Summary

Volume 289 presents the proceedings from the 7th Symposium on Advances in Approximate Bayesian Inference, held on 29 April 2025, at NTU College of Computing and Data Science, Singapore. Edited by James Urquhart Allingham and Siddharth Swaroop, this collection features seven papers exploring diverse methodologies within the field. Contributions include "Deep Q-Exponential Processes," "Massively Parallel Expectation Maximization For Approximate Posteriors," and an empirical study on "conformal prediction methods for in-context learning." Further research covers "Normalizing Flow Regression for Bayesian Inference with Offline Likelihood Evaluations," "$U$-ensembles" for improved diversity with unlabeled data, "Divide, Conquer, Combine Bayesian Decision Tree Sampling," and "Sparse Gaussian Neural Processes." These papers collectively advance techniques for handling complex probabilistic models and uncertainty quantification.

Key takeaway

For AI Scientists and Research Scientists focused on probabilistic modeling, this symposium volume offers a critical overview of current approximate Bayesian inference advancements. You should review these proceedings to identify novel techniques like massively parallel Expectation Maximization or normalizing flow regression that could enhance your model's scalability or uncertainty quantification. Consider exploring conformal prediction for in-context learning to improve confidence interval generation in your predictive systems.

Topics

• Approximate Bayesian Inference
• Deep Q-Exponential Processes
• Expectation Maximization
• Conformal Prediction
• Normalizing Flow Regression
• Bayesian Decision Trees

Best for: AI Scientist, Research Scientist

Related on AIssential

• v266: Proceedings of COPA 2025 — Conformal Prediction provides statistically rigorous uncertainty quantification across diverse machine learning applications and challenges.
• Probabilistic ML - 25 - Revision — Probabilistic inference, though often intractable, provides a foundational framework for machine learning through structured approximations.
• The Bayesian Workflow book is coming! — The "Bayesian Workflow" book offers practical guidance for applying Bayesian methods, addressing common challenges in statistical and computational practice.
• v271: Proceedings of ProbNum 2025 — Probabilistic Numerics integrates uncertainty quantification into numerical algorithms for robust computation.
• Tokenised Flow Matching for Hierarchical Simulation Based Inference — Likelihood factorisation and tokenised flow matching significantly reduce simulation costs in hierarchical SBI.
• Concentration and Calibration in Predictive Bayesian Inference — Predictive Bayesian Inference calibration hinges entirely on the accuracy of its forward predictive model.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Proceedings of Machine Learning Research.