Alexander Terenin

Project page: geospsr.github.io
Paper link: arxiv.org/abs/2503.19136
Link to my student's tweets on this work: x.com/sholalkere/s...

At ICML, we're presenting a paper on uncertainty-aware surface reconstruction!

Compared to previous approaches, we are able to completely remove the need for recursive linear solves for reconstruction and interpolation, using geometric GP machinery.

Check it out!

Virtual Seminar Series on Bayesian Decision-making and Uncertainty

Check out all of this season's seminars here: gp-seminar-series.github.io

This week’s virtual seminar on Bayesian Decision-making and Uncertainty is happening now!

Noémie Jaquier (KTH Royal Institute of Technology)
On Riemannian Latent Variable Models and Pullback Metrics

Livestream link: www.youtube.com/watch?v=61Be...

An Adversarial Analysis of Thompson Sampling for Full-information Online Learning: from Finite to Infinite Action Spaces We develop a form Thompson sampling for online learning under full feedback - also known as prediction with expert advice - where the learner's prior is defined over the space of an adversary's future...

If you're interested in this work, check out the updated preprint! It's got a lot of new stuff!

Link: arxiv.org/abs/2502.14790

We obtain a very simple condition which relates the prior covariance kernel with the adversary's function class in an easy-to-verify way that works in the bounded Lipschitz case.

I am very interested in extensions to more-general smoothness classes, and have ideas. Stay tuned!

This stands in contrast with prior arguments, which are linear-algebraic in flavor, involve bounding certain matrix norms by certain traces, and essentially-require independence in order to give sharp rates.

Specifically, we have a novel probabilistic argument which bounds Hessian-type terms which appear in the regret analysis of Gaussian follow-the-perturbed-leader algorithms, of which Thompson sampling is a special case.

Our argument works even with correlations!

I'm really excited about this! We had previously submitted this work to COLT, but it got rejected primarily for having "not enough new algorithmic results".

This is no longer a weakness of the paper. Our results in d>1 are all new!

The argument to get them is new as well!

Using this viewpoint, we introduce a Thompson sampling variant with a Gaussian process prior, and prove an adversarial guarantee against a bounded Lipschitz adversary.

So what's new?

We now have an analysis that works in any dimension!

The usual algorithms for this problem are convex-analytic - mirror descent variants and similar.

We develop a completely different looking, Bayesian way of thinking about what is going on - leading to Thompson sampling variants.

Let's first remember what this paper does: it gives a new way to think about designing algorithms for the so-called general-action-space online learning game - where a learner makes a prediction, and an adversary responds with a reward function in some class of possible rewards.

An Adversarial Analysis of Thompson Sampling for Full-information Online Learning: from Finite to Infinite Action Spaces We develop a form Thompson sampling for online learning under full feedback - also known as prediction with expert advice - where the learner's prior is defined over the space of an adversary's future...

And the link on arXiv: arxiv.org/abs/2502.14790

But first, the previous announcement of this work, which describes in more detail what is going on and why I think it's important: bsky.app/profile/avt....

This work is joint with Jeff Negrea, who has been a great pleasure to collaborate with!

We've got a major update to our preprint on adversarial regret guarantees for Thompson sampling!

As before, I think this is one of the most important projects I've worked on due to new algorithmic primitives that it - in principle - unlocks.

Thread below on what's new!

We’d like to announce next week’s virtual seminar on Bayesian Decision-making and Uncertainty!

Noémie Jaquier (KTH Royal Institute of Technology)
On Riemannian Latent Variable Models and Pullback Metrics

Sign-up link: gp-seminar-series.github.io

Virtual Seminar Series on Bayesian Decision-making and Uncertainty

Sign up for future seminars here: gp-seminar-series.github.io

This week’s virtual seminar on Bayesian Decision-making and Uncertainty is happening now!

Marvin Pförtner (University of Tübingen)
Computation-Aware Kalman Filtering and Smoothing

YouTube livestream: www.youtube.com/watch?v=0tG2...

We’d like to announce next week’s virtual seminar on Bayesian Decision-making and Uncertainty!

Marvin Pförtner (University of Tübingen)
Computation-Aware Kalman Filtering and Smoothing

Sign-up link: gp-seminar-series.github.io

Virtual Seminar Series on Bayesian Decision-making and Uncertainty

Sign up for future seminars here: gp-seminar-series.github.io

This week’s virtual seminar on Bayesian Decision-making and Uncertainty is happening now!

Yingzhen Li (Imperial College London)
On "Modernising" Sparse Gaussian Processes

YouTube livestream: www.youtube.com/watch?v=VbGW...

Virtual Seminar Series on Bayesian Decision-making and Uncertainty

gp-seminar-series.github.io

We’d like to announce next week’s virtual seminar on Bayesian Decision-making and Uncertainty!

Yingzhen Li (Imperial College London)
On Modernising Sparse Gaussian Processes

Sign-up link below!

Virtual Seminar Series on Bayesian Decision-making and Uncertainty

Sign up for future seminars here: gp-seminar-series.github.io

This week’s virtual seminar on Bayesian Decision-making and Uncertainty is happening now!

Sebastian Ament (Meta)
Unexpected Improvements to Expected Improvement for Bayesian Optimization

YouTube livestream: www.youtube.com/watch?v=B71O...

We’d like to announce next week’s virtual seminar on Bayesian Decision-making and Uncertainty!

Sebastian Ament (Meta)
Unexpected Improvements to Expected Improvement for Bayesian Optimization

Sign-up link: gp-seminar-series.github.io

Virtual Seminar Series on Bayesian Decision-making and Uncertainty

Sign up for future seminars here: gp-seminar-series.github.io

This week’s virtual seminar on Bayesian Decision-making and Uncertainty is happening now!

Paul Jensen (University of Michigan)
Exploring phenotypes and genotypes with a robot scientist

YouTube livestream: www.youtube.com/watch?v=UJQ2...

We’d like to announce next week’s virtual seminar on Bayesian Decision-making and Uncertainty!

Paul Jensen (University of Michigan)
Exploring phenotypes and genotypes with a robot scientist

Sign-up link below!

gp-seminar-series.github.io

PSA: If your GitHub Pro account expires - for instance due to a slow edu discount renewal - your GitHub Pages will go offline ***and might redirect to a SPAM website***!

I have no idea technically how.

If you land in this situation, move your DNS off GH pages immediately!