Ing Ren Tsang
@indentsang
Followers
169
Following
92
Media
4
Statuses
206
Joined December 2009
Ilya Sutskever's top 30 must-read research papers “If you really learn all of these, you’ll know 90% of what matters today” - Ilya Sutskever'
52
571
5K
O Centro de Informática (CIn) da UFPE abre inscrições para novo curso gratuito de capacitação na área de Fotografia Computacional com ênfase em Inteligência Artificial e Dispositivos móveis. https://t.co/szrhtf26NO
#CInUFPE
0
1
5
Seminário de Pesquisa do CIn-UFPE recebe os professores Luciano Barbosa, Teresa Ludermir e Tsang Ing Ren, para mesa-redonda com o tema “Chat-GPT: histórico, IA e pesquisa” 📢 Mais informações: https://t.co/Gq2Wf2DiWA
#CInUFPE
0
3
9
Out today! This Analysis describes the most recent Cell Tracking Challenge. Learn about the best tools for diverse tracking applications and what's still needed! https://t.co/ltHeq79UjO
0
21
59
Fusion power: how close are we? | FT Film https://t.co/pcsJAoiC72 via @FT
0
0
0
(1/2) 𝗪𝗵𝗲𝗻 𝗡𝗲𝗥𝗙 𝗺𝗲𝗲𝘁𝘀 𝘁𝗵𝗲 𝗠𝗲𝘁𝗮𝘃𝗲𝗿𝘀𝗲. Do you want to help to make 3D content creation a more believable and immerse metaverse experience?. Join us in the xRNeRF CVPR-2023 workshop, https://t.co/XZaaUVa15V
@akanazawa @jon_barron @daeil @aayush382 ,
sites.google.com
CVPR Workshop 2024 June 18, 2024 8:30 AM Summit 332, Seattle Convention Center, Seattle
1
23
113
🚀Excited+nervous to share our latest work on understanding geometric GNNs for biomolecules, materials, etc. "On the Expressive Power of Geometric GNNs" with @crisbodnar @SimMat20 @TacoCohen @pl219_Cambridge PDF: https://t.co/ywQCmMY0nG Code: https://t.co/h4soieuW1O Findings👇
7
96
423
Someone sent me AI art of Salmon in the river but HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAAHAHAHAHAHAHAHAH 💀💀💀
666
34K
268K
Dilated local attention: best of both worlds for scalable image segmentation.
Attention can now be applied in sliding-windows efficiently and flexibly with NA/DiNA, similar to what (dilated) Convolutions do to FC layers. They preserve locality, capture global context, and expand RFs exponentially, all with linear complexity! PwCode: https://t.co/yL3wjYPZWQ
8
68
432
It's a good day for text-to-image... 1) Make-A-Video ( https://t.co/c3tK4EYPBa): text-to-video w/o paired examples 2) Phenaki ( https://t.co/lzznehZ6nV): text-to-video (up to a few mins long (!)), prompts can change over time. 3) DreamFusion ( https://t.co/IUNAD59nvM): text-to-3D
6
94
484
What does last authorship signify to you in a (non-alphabetically authored) machine learning paper? What did that person do?
21
2
42
First Principles of Computer Vision, Columbia University Really nice lectures on the physical and mathematical foundations of computer vision. 140 videos that you can watch at your pace. Slides are also provided to follow along.
19
367
2K
It turns out that *all* independently trained neural nets form a connected, multidimensional manifold of low loss- you can always form a low-loss path from one SGD solution to any other. This can be used for efficient generation of ensembles.
arxiv.org
With a better understanding of the loss surfaces for multilayer networks, we can build more robust and accurate training procedures. Recently it was discovered that independently trained SGD...
17
64
530
Neural Tangent Kernels (NTKs) are used to analyze generalization of #neuralnetworks, but usually in unrealistic scenarios. My student @mselezniova got her paper on "NTKs Beyond the Infinite-Width Limit" accepted by @icmlconf 2022 as a spotlight: https://t.co/LOYH1cMHOf. Congrats!
4
14
123
