Found a kindred spirit at #icml2023 #feelthelearn

Hahahaha #ICML2023

The code for E(n) Equivariant Simplicial Message Passing Networks (ICML, 2023) has been released 🦦

Code: github.com/flooreijkelboo…
ArXiv: arxiv.org/abs/2305.07100

Thanks again to Rob and Erik Bekkers for the collab :D

Video lectures, ICML 2023 Tutorials
icml.cc/virtual/2023/e…

Learning under distribution shifts #icml #ICML2023

Best poster at ICML Poster Session
#ICML 2023
#ICML

Visit our poster on using simplicial complexes to describe complex geometry in graphs at #ICML2023 :

Tomorrow (27/7) 10:30, exhibit hall 1, poster 705!

Excited to share that our paper

“Adversarial Training Should Be Cast As a Non-Zero-Sum Game”

won the *𝐛𝐞𝐬𝐭 𝐩𝐚𝐩𝐞𝐫 𝐚𝐰𝐚𝐫𝐝* at the AdvML workshop at #ICML2023 ! 🚀

Paper: arxiv.org/abs/2306.11035
Talk: Friday at 10am in Ballroom A

Want to know more? Check out this 🧵

At #ICML2023 , Chay (Chaitanya (Chay) Ryali) will present Hiera, a hierarchical vision transformer that is fast, powerful, and simple. Code+models at: github.com/facebookresear…

If interested, please come to the oral presentation on Tue 25 Jul 5:30pm HST or poster #219 on Wed 26 Jul 2pm HST.

Optimizing latency & accuracy for neural networks that fit on your phone! Introducing UPSCALE - a new export algorithm for pruned vision models with a 2.1% accuracy boost and up to 2X faster inference speeds! #ICML2023 🧵👇
Website: upscale.aaalv.in

Causal effect of X on Y cannot be identified when X, Y are confounded by a latent variable Z. From p(x, y), one can bound p(y|do(x)) thanks to Tian and Pearl (2000).

But what if we know that the confounding is 'weak'?
#ICML2023

Can networks discover crucial geometric measurements like height/radius from binary task success signals? Check out our #ICML2023 paper on geometric eigen-length learning: yijiaweng.github.io/geo-eigen-leng….

Come by our poster at Exhibit Hall 1 #640 today at 10:30 am!

Wonderful dinner w/ Yaron Lipman Haggai Maron Francesco Di Giovanni Michael Bronstein Emanuele Rossi #ICML2023

AI&HCI workshop at ICML2023 is happening sites.google.com/view/aihci James Landay is giving the first keynote!

If you are in Hawaii for #ICML2023 , visit our poster on 'LEAD: Min-Max optimization from a physical perspective' at the #Frontiers4LCD workshop at 4pm, Ballroom B.
If not, you can find the paper, blogpost, code and a short video here: reyhaneaskari.github.io/LEAD.html

Noise2NoiseMapping
Learning Signed Distance Functions from Noisy 3D Point Clouds via Noise to Noise Mapping (ICML 2023 Oral)
github.com/mabaorui/Noise…

Current SOTA methods for scene understanding, though impressive, often fail to decompose out-of-distribution scenes.

In our #ICML2023 paper, Slot-TTA (slot-tta.github.io) we find that optimizing per test sample over reconstruction loss improves scene decomposition.

PaLM-E is a generalist, embodied language model for robotics. 🤖

It solves many tasks on 𝘮𝘶𝘭𝘵𝘪𝘱𝘭𝘦 types of robots and for 𝘮𝘶𝘭𝘵𝘪𝘱𝘭𝘦 modalities, including images and neural scene representations.

Hear more from researchers at #ICML2023 :
📍Booth #109
⌚10.30 HST

Excited to present the 🥮Flan Collection🥮 poster at #ICML2023 tomorrow.

Come say hello at Exhibition Hall 1, #130!

A highlight 🧵on Flan-PaLM 🌴, Flan-T5 and Flan data in research and production.

1/