AABI 2025

7th Symposium on Advances of Approximate Bayesian Inference

Co-located with ICLR 2025 in Singapore

Tuesday, 29 April 2025

Venue: NTU College of Computing and Data Science

Image by Jenty

Attendance Information

AABI 2025 has been concluded. See you at AABI 2026 next year!

Venue

AABI 2025 will be hosted at the NTU College of Computing and Data Science.

Lecture Theatre 1
76 Nanyang Drive
Level 2, North Spine Plaza
Nanyang Technological University
Singapore 637331

Schedule

Time Event Note
08:00 – 09:00 Check-In
09:00 – 09:13 Opening Remarks Tim G. J. Rudner and Vincent Fortuin
09:13 – 09:15 Sponsor Remarks BASIS

Session 1 — Chair: Christian Naesseth
09:15 – 10:00 Invited Talk 1 Adaptive Bayesian Intelligence (ABI)
Mohammad Emtiyaz Khan
10:00 – 10:15 Contributed Talk 1 Sparse Gaussian Neural Processes
Tommy Rochussen, Vincent Fortuin
10:15 – 10:30 Contributed Talk 2 From predictions to confidence intervals: an empirical study of conformal prediction methods for in-context learning
Zhe Huang, Simone Rossi, Rui Yuan, Thomas Hannagan
10:30 – 11:00 Break Refreshments provided

Session 2 — Chair: Agustinus Kristiadi
11:00 – 11:45 Invited Talk 2 Bayesian Inference for Invariant Feature Discovery from Multi-Environment Data
Luhuan Wu
11:45 – 12:00 Contributed Talk 3 Deep Q-Exponential Processes
Zhi Chang, Chukwudi Paul Obite, Shuang Zhou, Shiwei Lan
12:00 – 12:15 Contributed Talk 4 Normalizing Flow Regression for Bayesian Inference with Offline Likelihood Evaluations
Chengkun LI, Bobby Huggins, Petrus Mikkola, Luigi Acerbi
12:15 – 14:00 Lunch & Poster Session 1 Lunch provided
Proceedings: All papers
Workshop: Papers 17–33
Fast track: Papers 12–24
Refer to the accepted papers list

Session 3 — Chair: James Allingham
14:00 – 14:45 Invited Talk 3 LLM Predictive Distributions Over Functions
James Requeima
14:45 – 15:00 Contributed Talk 5 Divide, Conquer, Combine Bayesian Decision Tree Sampling
Jodie Anne Cochrane, Adrian Wills, Sarah J Johnson
15:00 – 15:30 Break Refreshments provided
15:30 – 15:45 Contributed Talk 6 U-ensembles: Improved diversity in the small data regime using unlabeled data
Konstantinos Pitas, Hani Anouar Bourrous, Julyan Arbel
15:45 – 16:00 Contributed Talk 7 Massively Parallel Expectation Maximization For Approximate Posteriors
Thomas Heap, Sam Bowyer, Laurence Aitchison
16:00 – 17:00 Poster Session 2 Proceedings: All papers
Workshop: Papers 1–16
Fast track: Papers 1–11
Refer to the accepted papers list
17:00 – 17:45 Panel Discussion The Role of Probabilistic Machine Learning in the Age of Foundation Models and Agentic AI
Mohammad Emtiyaz Khan, Luhuan Wu, James Requeima
Moderator: Siddharth Swaroop
17:45 – 18:00 Closing Remarks Vincent Fortuin

Sponsors

We are grateful for the supports from our sponsors.


Platinum Sponsors


Silver Sponsors