Tim G. J. Rudner

CDS Faculty Fellow @timrudner.bsky.social served as general chair for the 7th Symposium on Advances in Approximate Bayesian Inference, held April alongside ICLR 2025.

The symposium explored connections between probabilistic machine learning and AI safety, NLP, RL, and AI for science.

Congratulations again!

Congratulations Umang!

AI in Military Decision Support: Balancing Capabilities with Risk CDS Faculty Fellow Tim G. J. Rudner and colleagues at CSET outline responsible practices for deploying AI in military decision-making.

CDS Faculty Fellow Tim G. J. Rudner (@timrudner.bsky.social) and colleagues at CSET — @emmyprobasco.bsky.social, @hlntnr.bsky.social, and Matthew Burtell — examine responsible AI deployment in military decision-making.

Read our post on their policy brief: nyudatascience.medium.com/ai-in-milita...

The result in this paper I'm most excited about:

We showed that planning in world model latent space allows successful zero-shot generalization to *new* tasks!

Project website: latent-planning.github.io

Paper: arxiv.org/abs/2502.14819

#1: Can Transformers Learn Full Bayesian Inference In Context? with @arikreuter.bsky.social @timrudner.bsky.social @vincefort.bsky.social

Very excited that our work (together with my PhD student @gbarto.bsky.social and our collaborator Dmitry Vetrov) was recognized with a Best Paper Award at #AABI2025!

#ML #SDE #Diffusion #GenAI 🤖🧠

Congratulations to the #AABI2025 Proceedings Track Best Paper Award recipients!

Congratulations to the #AABI2025 Workshop Track Outstanding Paper Award recipients!

We concluded #AABI2025 with a panel discussion on

**The Role of Probabilistic Machine Learning in the Age of Foundation Models and Agentic AI**

Thanks to Emtiyaz Khan, Luhuan Wu, and @jamesrequeima.bsky.social for participating!

.@jamesrequeima.bsky.social gave the third invited talk of the day at #AABI2025!

**LLM Processes**

Luhuan Wu is giving the second invited talk of the day at #AABI2025!

**Bayesian Inference for Invariant Feature Discovery from Multi-Environment Data**

Watch it on our livestream: timrudner.com/aabi2025!

Emtiyaz Khan is giving the first invited talk of the day at #AABI2025!

We just kicked off #AABI2025 at NTU in Singapore!

We're livestreaming the talks here: timrudner.com/aabi2025!

Schedule: approximateinference.org/schedule/

#ICLR2025 #ProbabilisticML

AABI 2025 · Luma 7th Symposium on Advances of Approximate Bayesian Inference (AABI) https://approximateinference.org/schedule

Make sure to get your tickets to #AABI2025 if you are in Singapore on April 29 (just after #ICLR2025) and interested in probabilistic ML, inference, and decision-making!

Tickets (free but limited!): lu.ma/5syzr79m
More info: approximateinference.org

#ProbabilisticML #Bayes #UQ #ICLR2025 #AABI2025

Make sure to get your tickets to AABI if you are in Singapore on April 29 (just after #ICLR2025) and interested in probabilistic modeling, inference, and decision-making!

Tickets (free but limited!): lu.ma/5syzr79m
More info: approximateinference.org

#Bayes #MachineLearning #ICLR2025 #AABI2025

Putting Explainable AI to the Test: A Critical Look at AI Evaluation Approaches | Center for Security and Emerging Technology Explainability and interpretability are often cited as key characteristics of trustworthy AI systems, but it is unclear how they are evaluated in practice. This report examines how researchers evaluat...

CDS Faculty Fellow @timrudner.bsky.social, with @minanrn.bsky.social & Christian Schoeberl, analyzed AI explainability evals, finding a focus on system correctness over real-world effectiveness. They call for the creation of standards for AI safety evaluations.

cset.georgetown.edu/publication/...

How the U.S. Public and AI Experts View Artificial Intelligence These groups are far apart in their enthusiasm and predictions for AI, but both want more personal control and worry about too little regulation.

A great Pew Research survey:

"How the U.S. Public and AI Experts View Artificial Intelligence"

Everyone working in ML should read this and ask themselves why experts and non-experts have such divergent views about the potential of AI to have a positive impact.

www.pewresearch.org/internet/202...

This is an excellent article!

Steering foundation models towards trustworthy behaviors is one of the most important research directions today.

Helen is a deep and rigorous thinker, and you should definitely subscribe to her Substack!

I'm super excited to see our #CSET report on **AI-enabled military decision support systems** being released today!

Great work by @emmyprobasco.bsky.social, @hlntnr.bsky.social, and Matthew Burtell!

Putting Explainable AI to the Test: A Critical Look at AI Evaluation Approaches | Center for Security and Emerging Technology Explainability and interpretability are often cited as key characteristics of trustworthy AI systems, but it is unclear how they are evaluated in practice. This report examines how researchers evaluat...

Evaluations of AI explainability claims are not clear cut.

@minanrn.bsky.social, @timrudner.bsky.social & Christian Schoeberl show that in the domain of recommender systems – where explanations are key – there are different notions of what explainability means: cset.georgetown.edu/publication/...

Putting Explainable AI to the Test: A Critical Look at AI Evaluation Approaches | Center for Security and Emerging Technology Explainability and interpretability are often cited as key characteristics of trustworthy AI systems, but it is unclear how they are evaluated in practice. This report examines how researchers evaluat...

[1/6] Discourse around AI evaluations has focused a lot on testing LLMs for catastrophic risks. In a new @csetgeorgetown.bsky.social report, Christian Schoeberl, @timrudner.bsky.social, and I explore another side of AI evals: evals of claims about the trustworthiness of AI systems

Check out our paper on the quality of interpretability evaluations of recommender systems:

cset.georgetown.edu/publication/...

Led by @minanrn.bsky.social and Christian Schoeberl!

@csetgeorgetown.bsky.social

📣 Jobs alert

We’re hiring postdoc and research engineer to work on UQ for LLMs!! Details ⬇️

#ai #llm #uq

You still have a chance to submit your work to all tracks for AABI. New deadline is February 14 for both Workshop and Proceedings track!

#AABI2025 #ML #ICLR2025 #Stats #Bayes

Great news for everyone struggling with concurrent AABI and UAI submissions: We've extended the AABI deadline by one week, so there's no excuse not to submit anymore 😉

Call for Papers

We have extended the #AABI workshop and proceedings deadlines!

*New deadlines:*

Workshop Track: February 14, AoE
Proceedings Track: February 14, AoE
approximateinference.org/call/

#ProbML #AABI #ICLR #Bayes

We have extended the workshop and proceedings deadlines:

*New deadlines:*

Workshop Track: February 14, AoE
Proceedings Track: February 14, AoE

approximateinference.org/call/

#ProbML #AABI #ICLR

Home

We particularly welcome submissions that explore connections between probabilistic machine learning and other fields such as
- deep learning
- NLP
- active learning
- RL
- compression
- AI safety
- AI for scientific discovery
- causal inference

approximateinference.org

This year’s symposium will have an *expanded scope* and will be focused on the development, analysis, and application of *probabilistic machine learning methods* broadly construed.

#AABI #ProbML #Bayes