Frustrated that your ostriches are never behaving? This is now a thing of the past! We open-source the code for our notable-top-25% ICLR2023 paper DEP-RL (openreview.net/forum?id=C-xa_…)
Now everyone can train their favorite musculoskeletal models with RL!

github.com/martius-lab/de…

Bispectral Neural Networks go to #ICLR2023 !

In this work, we present a new neural network architecture capable of learning unknown groups purely from the symmetries implicit in data

—with Christian Shewmake, Bruno Olshausen, and Christopher Hillar

1/17

I am excited to share “3D generation on ImageNet”: our recent work on 3D synthesis, which got accepted to #ICLR2023 with the Top-5% (Oral) designation!

snap-research.github.io/3dgp/

Excited to present at #iclr2023 ! Our new fluid mechanics dataset, EAGLE, is a game changer with over 1.1 million super complex mesh simulations, streamlining future deep fluid simulators 🌊!

Project : eagle-dataset.github.io
Paper : openreview.net/forum?id=mfIX4…

Local SGD, though designed to reduce communication, can generalize better than SGD! Our #ICLR2023 paper gives the first theoretical explanation of this phenomenon: local steps inject extra noise, driving the iterate to drift faster to flatter minima on the minimizer manifold. 1/4

#Differentiating through discrete optimization solvers can be so easy!
They can often be treated as an Identity function on the backward pass. See our #ICLR2023 paper: openreview.net/forum?id=JZMR7…
Previous work, such as #Blackboxbackprop , requires an additional call to the solver.🧵

NeRF-SOS is a self-supervised framework for object segmentation using Neural Radiance Fields. It leverages a novel contrastive loss to achieve superior performance without explicit 3D supervision.[ICLR2023]

github.com/VITA-Group/NeR…

I’m excited to share our latest #ICLR2023 paper

🏎️ MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learning 🏎

Paper: arxiv.org/abs/2303.03376
Website: maestro.samvelyan.com

Highlights: 👇

1/ Thrilled to share our #ICLR2023 paper on Proto-Value Networks (PVN) now on arXiv!

Through a collection of self-supervised tasks, PVNs learn to capture the spatiotemporal structure of the environment, resulting in state-of-the-art representations!👇

arxiv.org/abs/2304.12567

Facing an equivariant task with mismatched image & object transformations? Our #ICLR2023 paper discovers the power of equivariant network when enforcing symmetry to out-of-distribution data. Join my talk tmr (May 1st) in Oral2 Track4 (AD10) at 3:10 or drop by Poster Session 2! 1/

Thrilled to share new state-of-the-art 3D molecule generation results with our new Geometry-Complete Diffusion Model (GCDM). I'll be presenting GCDM tomorrow at #ICLR2023 MLDD. Look forward to seeing you there!
Paper: arxiv.org/abs/2302.04313
Code: github.com/BioinfoMachine…

#ICLR2023 Ever wondered when can robots learn to make latte art? Please check out FluidLab, which took the first step towards unlocking new capabilities of robotic dexterity to handle various kinds of fluids!

Project page: fluidlab2023.github.io
Code: github.com/zhouxian/Fluid…

[1/8] How does the brain develop the ability to predict the next frame of a movie?

In an #ICLR2023 paper [w/ Elias Issa], we showed the link between robustness to noise on input images and the temporal straightening of movies.

openreview.net/forum?id=mCmer…

Today at #ICLR2023 , I’m excited to present our paper: “Investigating Multi-task Pretraining and Generalization in Reinforcement Learning”.

Interested in models that can learn quickly by observing only a few examples in a multimodal space? In our #ICLR2023 paper on multimodal few-shot learning, we leverage existing large-scale VL models to learn shared meta-knowledge and adapt rapidly to new samples! 🥳🧵⏭️ 1/n

🎾🥳DEEP LEARNING GRAND SLAM🥳🎾
(get a paper accepted at ICLR, ICML, NeurIPS within one year 😜)

ICLR2023: [MHNfs](tinyurl.com/2jjuv35z)
ICML2023: [CLAMP](tinyurl.com/2ekkss2y)
NeurIPS2023: Initialization for Input-Convex Nets

Credits should go to the PhD students 🎓👏

Generative models are awesome at producing data, and weak supervision is great at efficient labeling. Can we combine them to get cheap datasets for training or fine-tuning?

Excited to present our #ICLR2023 paper 'Generative Modeling Helps Weak Supervision (and Vice Versa)'

Excited to present our award-winning DreamFusion research today at #ICLR2023 ! Talk at 3:40pm in AD12, and poster #73 at 4:30pm. Have a few souvenirs to distribute too 🐸👻🐷🐶

LLMが人のウェブ上・映像上の「行動」を学習し、理解し、予測できるようにするアーキテクチャ『LCBM（大規模コンテンツ行動モデル）』が開発されました。
Adobeなどの研究グループによってICLR2023で発表されています。

○ Ashmit Khandelwal et al. Large Content And Behavior Models To…

Happy to share that ProNet is accepted to #ICLR2023 ! ProNet learns hierarchical representations of proteins. It is effective and efficient (achieving complete representations at each level) and can be used for a variety of protein-related tasks!
openreview.net/pdf?id=9X-hgLD…