🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Vision transformers (ViT) and MLP-Mixer models tend to produce sharp loss landscapes. Thus, they stand to benefit most from the recently proposed sharpness-aware minimization (SAM) optimizer! (see Foret et al. 2021, ICLR for more on SAM) arxiv.org/abs/2106.01548

Contrastive learning aims to learn representation such that similar samples stay close, while dissimilar ones are far apart. It can be applied to supervised / unsupervised data and has been shown to achieve good results on various tasks. 📚 A long read: lilianweng.github.io/lil-log/2021/0…

The submission system for the NeurIPS 2021 workshop on ImageNet is open: openreview.net/group?id=NeurI… The submission deadline for all three tracks are 18 September 2021, 23:59 AOE Visit the workshop web page sites.google.com/view/imagenet-… for more information!

If you've been working with ImageNet and want to share interesting results that can not be summarized by "beating some SoTA", you should consider submitting to the NeurIPS workshop on ImageNet: Past, Present, and Future. Deadline Sept 18th. openreview.net/group?id=NeurI…

Announcing 📢 registration for the '2nd NAVER LABS Europe International AI for Robotics workshop' 29-30 Nov 📢 (Virtual) Star lineup of speakers! Best poster award 🏆🏆🏆 Registration FREE - Please retweet & share!!! #AI4Robotics Program & registration: bit.ly/3jKhOyP

PyTorch Adapt is a new modular library for domain adaptation. Intro post: medium.com/@tkm45/deep-do… GitHub: github.com/KevinMusgrave/… Jupyter notebooks: github.com/KevinMusgrave/… Installation: pip install pytorch-adapt #PyTorch #MachineLearning #DeepLearning

This will be a fascinating workshop! Want to learn more before you go? Read the preview on Computer Vision News with Olga Russakovsky ==> rsipvision.com/ComputerVision… NAVER LABS Europe Diane Larlus #NeurIPS2021 RSIP Vision

In about 3 hours, we going LIVE on the #NeurIPS2021 workshop on ImageNet: past, present and future. Come check the amazing talks, panel and poster sessions! You can check the schedule 👇 or here: sites.google.com/view/imagenet-…

First session is starting NOW!

After three great talks, our first panel on "the future of ImageNet" is about to start w Olga Russakovsky Kaiyu Yang Vittorio Ferrari Matthias Bethge #NeurIPS2021 ImageNet: past, present, future

Oops I forgot announcing the second track! "Are we done with ImageNet?" by Alexander Kolesnikov is now live and other talks by Becca Roelofs Is ImageNet Solved? Evaluating Machine Accuracy Shibani Santurkar From ImageNet to Image Classification are also available x.com/SanghyukChun/s…

3rd session of our WS has begun Sharon Y. Li Uncovering the Deep Unknowns of ImageNet Model: Challenges and Opportunties Ross Wightman ImageNet models from the trenches Dawn Song Dan Hendrycks Using ImageNet to Measure Robustness and Uncertainty x.com/SanghyukChun/s…

We finally have the final (live) invited talk today. ImageNets of "x": ImageNet's Infrastructural Impact by Remi Denton and Alex Hanna (اليكس حنٌا) See you at the live Zoom! neurips.cc/virtual/2021/w… x.com/SanghyukChun/s…

Vision Transformers (ViTs) are an interesting development in computer vision. One downside: they are more data hungry than CNN. Presumably because they have less of a spatial / locality inductive bias so they require more data to obtain acceptable visual representations 🧵 [1/7]

You can start using FFCV today: check out the repo (github.com/MadryLab/ffcv) and docs (docs.ffcv.io)---we even have a Slack! Stay tuned for a blog post, and a paper explaining the details. w/ Guillaume Leclerc Andrew Ilyas Logan Engstrom Sam Park @hadisalmanx (3/3)

Happy to finally announce the release of our models! Improvements due to the data from: x.com/Nils_Reimers/s… and the PLM from arxiv.org/abs/2108.05540. We also added the needed code to index and retrieve using Anserini!

A recurring mis-conception in ML is that setting standards for ethical research practice is the same as censorship. It's not - it's about meeting community norms / expectations. This would be like saying a reviewer's request for reporting ablations or uncertainty was censorship.

Happy to be recognized as an outstanding reviewers for #ECCV2022

Interested in Cross-Modal retrieval? Feel that evaluation on caption datasets is limited and noisy? The AToMiC@TREC2023 task is here: a real, large scale problem (retrieve illustration for Wikipedia sections), with human judgements. Deadline in 45 days! Krishna Srinivasan Jimmy Lin

Multi-HMR has been accepted to #ECCV2024🤸. With Multi-HMR, we won the ROBIN challenge on 3D human reconstruction at #CVPR2024. Today, we're releasing the training code on top of the existing demo code: github.com/naver/multi-hmr See you in Milano 👋