Robert A. Vandermeulen

conjunction *with the* first

🚨 Local dependence → manifold hypothesis Distant independence → Markov random fields New preprint: the latter works in conjunction first, massivly improves data efficiency, and can be exploited by DNNs arxiv.org/abs/2411.15095 With @itsrainingdata sites.google.com/view/wmtai

As promised: arxiv.org/abs/2410.07685

🎉I'm happy to announce that our paper "Breaking the curse of dimensionality in structured density estimation" was accepted to NeurIPS 2024. 🎉 with: Wai Ming Tai (Nanyang Tech. Uni. - sites.google.com/view/wmtai ) and @itsrainingdata (U Chicago) -Preprint forthcoming-

NeurIPS preprint: arxiv.org/abs/2410.07685 Standard models for spatial, sequential, hierarchal, etc data result in drastically reduced effective dimension for density estimation. This effective dimension is orthogonal to other approaches such as sparsity or manifold hypothesis.

Just noticed that Tsybakov (author of one of my favorite textbooks: Intro. to Nonparametric Estimation) wrote a nice extension of my 2021 NeurIPS paper "Beyond Smoothness: Incorporating Low-Rank Analysis into Nonparametric Density Estimation." Exciting! hal.science/hal-04557030/

Congratulations: The paper “Set Learning for Accurate and Calibrated Models” is accepted at the #ICLR2024. Authors: @lukas_mut, @robvdm, Q. Zhang, T. Unterthiner, K.-R. Müller. Preprint: arxiv.org/abs/2307.02245 Wrap up: t1p.de/ijaiw @GoogleDeepMind @ml_tuberlin

OKO got accepted to #ICLR2024🎉🥳 OKO is very simple to use: Just change your sampling from single data point sampling to set sampling and compute cross-entropy for the sum of set data point logits rather than for single data point logits. You will get much better calibration! 🔥

Fun work between @ml_tuberlin @bifoldberlin @GoogleDeepMind 🎆 Thanks to my co-authors @robvdm @TomUnterthiner @richard_qiuyi_z and Klaus-Robert Müller!

🔥The CfP for Re-Align is up 🔥 If you are broadly interested in the alignment of two or more representation spaces, consider submitting to our workshop at #ICLR2024 and join us in beautiful Vienna!

Asking a famous researcher in field X about X. youtu.be/ZDNgmdGMpuY?si…

There are postdoc positions available at my research institute!

@vivekt17 @aus_hater @whstancil @kpanyc People should really source claims on Twitter. Would fix a lot of problems.

@aus_hater @whstancil @kpanyc Right! Would be great if this reason were hinted at or provided. Theres nothing to dig into with rich and well thought feedback like “pigheaded.”

one of the most annoying things on here is when you make a pretty elaborate argument someone happens to disagree with and they’re like “oh I get it now: you’re an idiot who is allergic to thinking, I should never have engaged”

@pronounced_kyle I bought a pack of four of these books for my son. It’s a cute idea but in all honesty, they are not very good, even for babies.

THIS BOOK IS A TRAVESTY. Someone just bought it for my soon-to-be-born son. Here is a replay of my internal monologue as I gave it a first read:

@oneunderscore__ Is this real?

We are reaching terminal levels of divorce.

At the poster for @lukas_mut's paper "Improving neural network representations using human similarity judgments" twitter.com/AndrewLampinen… Wed 13 Dec 10:45 a.m. CST — 12:45 p.m. CST Great Hall & Hall B1+B2 (level 1) #300 nips.cc/virtual/2023/p…

It seems like all math tests for LLMs involve just the direct application of various rules, like a bunch of trig or calc rules, to solve the problem. How about getting a LLM to prove the intermediate value theorem or some other proof that requires a clever construction?

@s_mandt I’m not familiar with this. Is this for PhD or masters admissions? How common is it?

This tweet has hit a nerve. When will graduate admission offices remove useless hurdles when submitting letters?

@MaxPaxCat @broncoslyfer @whstancil Trump refused to turn them over though.

@broncoslyfer @whstancil Trump's facing federal charges for doing the same thing with classified docs as Biden. They're using the criminal justice system to try to manipulate elections, and you can't see that?

one reason biden is sunk is that 95% of you are utter moral cowards who are so afraid being deemed uncool by your internet friends that you can't find it in yourself to choke out the admission that maybe trump is worse and voting against biden is really dangerous