looking at you #AISTATS2025

Out-of-distribution robustness for multivariate analysis via causal regularisation

🙌Some days ago our colleague Homer Durand presented his work "Out-of-distribution robustness for multivariate analysis via causal regularisation" at the #AISTATS2025.

🔗 Read the full paper here: tr.ee/Out-of-DistributionRobustness

Back from #AISTATS2025! Had a lovely time there, I especially liked meeting our Journal-to-Conference track presenters in person (I might have snapped a picture or two 📸)

Already looking forward to the next edition of AISTATS… and of its Journal-to-Conference track with @pierrealquier.bsky.social!

Illustrative example of WS-GPLVM. This image shows how the model can retrieve the component mixtures, latent variables and pure spectra from a data set where some spectra have known weight fractions and some don't.

New paper at #AISTATS2025: Weighted-Sum of Gaussian Process Latent Variable Models (WS-GPLVM)
We tackle a core challenge in signal separation where pure components vary nonlinearly across samples using latent variable Gaussian processes. (1/8)

Super happy #xCEBRA was presented as a peer-reviewed proceedings paper today at #AISTATS2025 by @stes.io et al! This extends #CEBRA with the ability to attribute specific inputs to latents over time 🙌

📝: openreview.net/forum?id=aGr...
Code: github.com/AdaptiveMoto...
Demo: cebra.ai/docs/demo_no...

All or None: Identifiable Linear Properties of Next-Token... We analyze identifiability as a possible explanation for the ubiquity of linear properties across language models, such as the vector difference between the representations of “easy” and “easiest”...

Now in Thailand to present our paper at #AISTATS2025 🇹🇭🌴

📍Today at 3:00-6:00 pm, poster number 118!

More details here:

openreview.net/forum?id=XCm...

#AISTATS2025 happening in Phuket, Thailand! I have two papers at the conference:

1. Max-Rank: Efficient Multiple Testing for Conformal Prediction

2. Variational Combinatorial Sequential Monte Carlo for Bayesian Phylogenetics in Hyperbolic Space

Both at poster session 2!

If you are at #AISTATS2025 and are interested in concept erasure, talk to @somnathbrc.bsky.social at Poster Session 1 on Saturday May 3.

Quantifying Knowledge Distillation Using Partial Information Decomposition Knowledge distillation deploys complex machine learning models in resource-constrained environments by training a smaller student model to emulate internal representations of a complex teacher model. ...

📢 Knowledge distillation trains smaller student models from complex teacher models. But are all teachers equally helpful? Can we formally quantify useful distillable knowledge? Our paper at #AISTATS2025 explains distillation using Partial Information Decomposition. arxiv.org/abs/2411.07483

Excited to share our group’s latest work at #AISTATS2025! 🎓
Tackling concentration in dependent data settings with empirical Bernstein bounds for Hilbert space-valued processes.
📍Catch the poster tomorrow!

🔁 See the original tweet for details!

🚨 Poster at #AISTATS2025 tomorrow!
📍Poster Session 1 #125

We present a new empirical Bernstein inequality for Hilbert space-valued random processes—relevant for dependent, even non-stationary data.

w/ Andreas Maurer, @vladimir-slk.bsky.social & M. Pontil

📄 Paper: openreview.net/forum?id=a0E...

DDEQs: Distributional Deep Equilibrium Models through Wasserstein... Deep Equilibrium Models (DEQs) are a class of implicit neural networks that solve for a fixed point of a neural network in their forward pass. Traditionally, DEQs take sequences as inputs, but have...

If you're at #AISTATS2025, check out the presentation by Jonathan Geuter, in collaboration with Clément Bonet, @akorba.bsky.social and @dmelis.bsky.social.

'DDEQs: Distributional Deep Equilibrium Models through Wasserstein Gradient Flows'

openreview.net/forum?id=rFf...

#AI #statistics #ML

Come check out our #AISTATS2025 poster Sunday Hall A—E 79.

Previously we showed how to fit a full cov Gaussian approx via “batch and (score) match.” Here we show how to make the update cheaper using a “patch” step that projects the update to one that is low rank + diagonal.

Spectral Representation for Causal Estimation with Hidden Confounders
at #AISTATS2025

A spectral method for causal effect estimation with hidden confounders, for instrumental variable and proxy causal learning
arxiv.org/abs/2407.10448

Haotian Sun, @antoine-mln.bsky.social, Tongzheng Ren, Bo Dai

Just arrived in Phuket for #AISTATS2025. Can't wait to present our poster (in tube) about SNAP 🫰 on day 2, Sunday! Come check it out and let's chat about scalable causal discovery!

Density Ratio-based Proxy Causal Learning Without Density Ratios 🤔

at #AISTATS2025

An alternative bridge function for proxy causal learning with hidden confounders.
arxiv.org/abs/2503.08371
Bozkurt, Deaner, @dimitrimeunier.bsky.social, Xu

@huangdaolang.bsky.social will be presenting our work tomorrow on pretrained/amortized transformers for all sorts of inference tasks, go say hi!
#AISTATS2025

Poster session 1
Place: Hall A-E
Time: Sat 3 May 3 p.m. — 6 p.m.
Poster number: 23

Come say hi at my #AISTATS2025 poster tomorrow!
📍 Poster Session I #83
🕒 Sat, May 3 at 3pm
Semiparametric Conformal Prediction uses copulas and the one-step estimator to build efficient conformal sets over multiple response variables. Joint work w/ Rob Tibshirani, @kyunghyuncho.bsky.social

New in ML Blog Post

Here's another simultaneous post for #AISTATS2025 by @khimya.bsky.social and @ml4x.bsky.social, with more of a professional focus. Always good to get multiple perspectives! 2/3 virtual.aistats.org/Conferences/...

Tips on How to Connect at Academic Conferences I was a kinda awkward teenager. If you are a CS researcher reading this post, then chances are, you were too. How to navigate social situations and make friends is not always intuitive, and has to …

I wrote a post on how to connect with people (i.e., make friends) at CS conferences. These events can be intimidating so here's some suggestions on how to navigate them

I'm late for #ICLR2025 #NAACL2025, but in time for #AISTATS2025 #ICML2025! 1/3
kamathematics.wordpress.com/2025/05/01/t...

On my way to Mai Khao🏝️ for #AISTATS2025, where I will present A Causal Framework for Evaluating Deferring Systems, a joint work with Filippo Palomba, Jose Alvarez and Salvatore Ruggieri.

1/5

For the first time, Journal-to-Conference papers have entered the schedule of #AISTATS2025 🎉

To celebrate here is a bingo card: should you be among the first to meet all of our amazing Journal-to-Conference presenters (I need proof!), I’ll buy you a drink... provided that you can find me too!

Arriving in Singapore this afternoon 🛬 I'll attend #ICLR2025, #AABI2025, and #AISTATS2025 together with many of my students and collaborators to present our 2 orals, 5 posters, and 14 workshop contributions 🚀

Feel free to drop by!

If you are also going to #AISTATS2025, @somnathbrc.bsky.social will be presenting 𝐩𝐞𝐫𝐟𝐞𝐜𝐭 𝐜𝐨𝐧𝐜𝐞𝐩𝐭 𝐞𝐫𝐚𝐬𝐮𝐫𝐞.

Somnath will be at ICLR 2025. Please catch him and talk to him if you are interested in concept erasure / unlearning / etc.
bsky.app/profile/somn...

At #AISTATS2025, I will be giving an "Oral" presentation of our work on "Implicit Diffusion"

arxiv.org/abs/2402.05468

I will be attending #ICLR2025 in Singapore and #AISTATS2025 in Mai Khao over the next two weeks.

Looking forward to meeting new people and learning about new things. Feel free to reach out if you want to talk about Google DeepMind.

𝐇𝐨𝐰 𝐜𝐚𝐧 𝐰𝐞 𝐩𝐞𝐫𝐟𝐞𝐜𝐭𝐥𝐲 𝐞𝐫𝐚𝐬𝐞 𝐜𝐨𝐧𝐜𝐞𝐩𝐭𝐬 𝐟𝐫𝐨𝐦 𝐋𝐋𝐌𝐬?

Our method, Perfect Erasure Functions (PEF), erases concepts perfectly from LLM representations. We analytically derive PEF w/o parameter estimation. PEFs achieve pareto optimal erasure-utility tradeoff backed w/ theoretical guarantees. #AISTATS2025 🧵

Thrilled to share that this work with Yichen Song and
@aldopacchiano.bsky.social got accepted for an oral presentation at AISTATS 2025!

Pure Exploration with Feedback Graphs arxiv.org/pdf/2503.07824

Code: github.com/rssalessio/P...

#AISTATS #AISTATS2025 #ML

KLabが国際学会AISTATS2025で優れた研究成果を発表 KLabの論文がAISTATS2025に採択。AI開発を加速する新たなモデル選択手法を提案し、ゲーム産業に革新をもたらします。

KLabが国際学会AISTATS2025で優れた研究成果を発表 #東京都 #港区 #AI開発 #KLab #AISTATS2025

KLabの論文がAISTATS2025に採択。AI開発を加速する新たなモデル選択手法を提案し、ゲーム産業に革新をもたらします。

Visit our poster soon at #AISTATS2025! 🚀

@istecnico.bsky.social @itnewspt.bsky.social @ellisunitlisbon.bsky.social

Made in @sardine-lab-it.bsky.social !