
Bringing powerful machine-learning systems to daily-life devices | Director: Dr. Yingyan Lin | GaTech CS | #ML #hardware #accelerators #greenAI
Atlanta, GA
Joined July 2022
- Tweets32
- Following45
- Followers49
- Likes8
Efficient and Intelligent Computing Lab retweeted
Racism directed at any group of people is unacceptable. This time, it targets Chinese students; next time, it could be another. Generalizing the (assumed) behavior of one individual to an entire group is neither professional nor scholarly.
Thank you @jeffdean for speaking up!
3. 𝗛𝗶𝗻𝘁-𝗔𝘂𝗴: 𝗗𝗿𝗮𝘄𝗶𝗻𝗴 𝗛𝗶𝗻𝘁𝘀 𝗳𝗿𝗼𝗺 𝗩𝗶𝘀𝗶𝗼𝗻 𝗙𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹𝘀 𝘁𝗼𝘄𝗮𝗿𝗱𝘀 𝗕𝗼𝗼𝘀𝘁𝗲𝗱 𝗙𝗲𝘄-𝘀𝗵𝗼𝘁 𝗣𝗮𝗿𝗮𝗺𝗲𝘁𝗲𝗿-𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗩𝗶𝗧 𝗧𝘂𝗻𝗶𝗻𝗴
youtu.be/Ben48mkV5JY
2. 𝗖𝗮𝘀𝘁𝗹𝗶𝗻𝗴-𝗩𝗶𝗧: 𝗖𝗼𝗺𝗽𝗿𝗲𝘀𝘀𝗶𝗻𝗴 𝗦𝗲𝗹𝗳-𝗔𝘁𝘁𝗲𝗻𝘁𝗶𝗼𝗻 𝘃𝗶𝗮 𝗦𝘄𝗶𝘁𝗰𝗵𝗶𝗻𝗴 𝗧𝗼𝘄𝗮𝗿𝗱𝘀 𝗟𝗶𝗻𝗲𝗮𝗿-𝗔𝗻𝗴𝘂𝗹𝗮𝗿 𝗔𝘁𝘁𝗲𝗻𝘁𝗶𝗼𝗻 𝗗𝘂𝗿𝗶𝗻𝗴 𝗩𝗶𝘀𝗶𝗼𝗻 𝗧𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗲𝗿 𝗜𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲
youtu.be/2x31-_xbi5U
1. 𝗔𝘂𝘁𝗼-𝗖𝗔𝗥𝗗: 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗮𝗻𝗱 𝗥𝗼𝗯𝘂𝘀𝘁 𝗖𝗼𝗱𝗲𝗰 𝗔𝘃𝗮𝘁𝗮𝗿 𝗗𝗿𝗶𝘃𝗶𝗻𝗴 𝗳𝗼𝗿 𝗥𝗲𝗮𝗹-𝘁𝗶𝗺𝗲 𝗠𝗼𝗯𝗶𝗹𝗲 𝗧𝗲𝗹𝗲𝗽𝗿𝗲𝘀𝗲𝗻𝗰𝗲
youtu.be/XJP5G4GqjzE
Our paper titled “An Investigation on Hardware-Aware Vision Transformer Scaling” was accepted by the Transactions on Embedded Computing Systems (dl.acm.org/journal/tecs)!
Congratulations to Chaojian Li!
Congratulations to @ranery1998 (Haoran You) for being selected by the College of Computing Awards Committee to receive the 𝗢𝘂𝘁𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 𝗚𝗿𝗮𝗱𝘂𝗮𝘁𝗲 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗔𝘀𝘀𝗶𝘀𝘁𝗮𝗻𝘁 𝗔𝘄𝗮𝗿𝗱!
eiclab.scs.gatech.edu/assets…
Two papers accepted to ISCA 2023 (iscaconf.org/isca2023/), Orlando, here we come! 🙂
1. Instant-3D: Instant Neural Radiance Fields Training Towards Real-Time AR/VR 3D Reconstruction
Sixu Li*, Chaojian Li*, Wenbo Zhu, Boyang (Tony) Yu, Yang Zhao, Cheng Wan, Haoran You, Huihong Shi, Yingyan (Celine) Lin
Three papers accepted to CVPR 2023 (lnkd.in/eN7nn895), Vancouver, here we come! 🙂
3. 𝗛𝗶𝗻𝘁-𝗔𝘂𝗴: 𝗗𝗿𝗮𝘄𝗶𝗻𝗴 𝗛𝗶𝗻𝘁𝘀 𝗳𝗿𝗼𝗺 𝗩𝗶𝘀𝗶𝗼𝗻 𝗙𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹𝘀 𝘁𝗼𝘄𝗮𝗿𝗱𝘀 𝗕𝗼𝗼𝘀𝘁𝗲𝗱 𝗙𝗲𝘄-𝘀𝗵𝗼𝘁 𝗣𝗮𝗿𝗮𝗺𝗲𝘁𝗲𝗿-𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗩𝗶𝗧 𝗧𝘂𝗻𝗶𝗻𝗴
Paper: arxiv.org/abs/2211.10526
Code: github.com/GATECH-EIC/Castli…
Project: haoranyou.com/castling-vit/
Haoran You*, Yunyang Xiong*, Xiaoliang Dai, Bichen Wu, Peizhao Zhang, Haoqi Fan, Peter Vajda, Yingyan (Celine) Lin
2. 𝗖𝗮𝘀𝘁𝗹𝗶𝗻𝗴-𝗩𝗶𝗧: 𝗖𝗼𝗺𝗽𝗿𝗲𝘀𝘀𝗶𝗻𝗴 𝗦𝗲𝗹𝗳-𝗔𝘁𝘁𝗲𝗻𝘁𝗶𝗼𝗻 𝘃𝗶𝗮 𝗦𝘄𝗶𝘁𝗰𝗵𝗶𝗻𝗴 𝗧𝗼𝘄𝗮𝗿𝗱𝘀 𝗟𝗶𝗻𝗲𝗮𝗿-𝗔𝗻𝗴𝘂𝗹𝗮𝗿 𝗔𝘁𝘁𝗲𝗻𝘁𝗶𝗼𝗻 𝗗𝘂𝗿𝗶𝗻𝗴 𝗩𝗶𝘀𝗶𝗼𝗻 𝗧𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗲𝗿 𝗜𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲
1. 𝗔𝘂𝘁𝗼-𝗖𝗔𝗥𝗗: 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗮𝗻𝗱 𝗥𝗼𝗯𝘂𝘀𝘁 𝗖𝗼𝗱𝗲𝗰 𝗔𝘃𝗮𝘁𝗮𝗿 𝗗𝗿𝗶𝘃𝗶𝗻𝗴 𝗳𝗼𝗿 𝗥𝗲𝗮𝗹-𝘁𝗶𝗺𝗲 𝗠𝗼𝗯𝗶𝗹𝗲 𝗧𝗲𝗹𝗲𝗽𝗿𝗲𝘀𝗲𝗻𝗰𝗲
Yonggan Fu, Yuecheng Li, Chenghui Li, Jason Saragih, Peizhao Zhang, Xiaoliang Dai, Yingyan (Celine) Lin
Efficient and Intelligent Computing Lab retweeted
Awesome website summarizing our latest TMLR paper demonstrating how deep networks pruning can be easily explained/visualized and improved simply by formulating it in terms of the DN's spline partition!
Paper: arxiv.org/abs/2101.02338
Code: github.com/GATECH-EIC/Spline…
We just built a website for this project 🤗:
haoranyou.com/spline-eb/
Three papers accepted to DAC 2023 (dac.com/), Moscone West, here we come! 🙂 !
3. 𝗜𝗻𝘀𝘁𝗮𝗻𝘁-𝗡𝗲𝗥𝗙: 𝗜𝗻𝘀𝘁𝗮𝗻𝘁 𝗢𝗻-𝗗𝗲𝘃𝗶𝗰𝗲 𝗡𝗲𝘂𝗿𝗮𝗹 𝗥𝗮𝗱𝗶𝗮𝗻𝗰𝗲 𝗙𝗶𝗲𝗹𝗱 𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴 𝘃𝗶𝗮 𝗔𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺-𝗔𝗰𝗰𝗲𝗹𝗲𝗿𝗮𝘁𝗼𝗿 𝗖𝗼-𝗗𝗲𝘀𝗶𝗴𝗻𝗲𝗱 𝗡𝗲𝗮𝗿-𝗠𝗲𝗺𝗼𝗿𝘆 𝗣𝗿𝗼𝗰𝗲𝘀𝘀𝗶𝗻𝗴



