Renjie Liao (@lrjconan) 's Twitter Profile
Renjie Liao

@lrjconan

Fullmetal Alchemist of Deep Learning

ID: 301341538

linkhttps://lrjconan.github.io/ calendar_today19-05-2011 09:23:57

110 Tweet

1,1K Followers

690 Following

ACM Education & Learning Center (@acmeducation) 's Twitter Profile Photo

December 7, join Raquel Urtasun, CEO Waabi, Professor U of T Department of Computer Science & co-founder Vector Institute, for the #ACMTechTalk "An AI-first Approach to Accelerating Autonomous Trucking." ACM Journal on Autonomous Transportation co-EiCs Satish Ukkusuri & Vaneet Aggarwal moderate. Register free: bit.ly/3XowQfM

December 7, join <a href="/RaquelUrtasun/">Raquel Urtasun</a>, CEO <a href="/Waabi_ai/">Waabi</a>, Professor <a href="/UofTCompSci/">U of T Department of Computer Science</a> &amp; co-founder <a href="/VectorInst/">Vector Institute</a>, for the #ACMTechTalk "An AI-first Approach to Accelerating Autonomous Trucking." <a href="/ACM_JATS/">ACM Journal on Autonomous Transportation</a> co-EiCs Satish Ukkusuri &amp; Vaneet Aggarwal moderate.

Register free: bit.ly/3XowQfM
Renjie Liao (@lrjconan) 's Twitter Profile Photo

I'm excited and honored to be appointed as the new Canada CIFAR AI Chair through #PCAIS! I’m looking forward to working with the Vector Institute & University of British Columbia community to grow Canada's robust AI research ecosystem + advance Canada’s global leadership in #AI

Mengye Ren (@mengyer) 's Twitter Profile Photo

Wondering about how to train deep neural networks without backprop? Check out our ICLR 2023 paper: arxiv.org/abs/2210.03310 Forward gradient computes gradient information from forward pass. But it is slow and noisy — it computes the directional gradient along a random weight

Paul Vicol (@paulvicol) 's Twitter Profile Photo

Had fun presenting ES-Single at #ICML2023! ES-Single is a method for low-variance, unbiased gradient estimation in unrolled computation graphs, that has lower variance than PES and outperforms it on a range of tasks. ICML: icml.cc/virtual/2023/p… Colab: tinyurl.com/es-single-colab

Had fun presenting ES-Single at #ICML2023! ES-Single is a method for low-variance, unbiased gradient estimation in unrolled computation graphs, that has lower variance than PES and outperforms it on a range of tasks.

ICML: icml.cc/virtual/2023/p…
Colab: tinyurl.com/es-single-colab
Renjie Liao (@lrjconan) 's Twitter Profile Photo

Yifan is a rising star in computer vision and such a nice and inspiring person. Justice must be done! An thorough investigation by ETH Zurich is needed!

Renjie Liao (@lrjconan) 's Twitter Profile Photo

Welcome to submit your work to NeurIPS 2024 workshop on Adaptive Foundation Models! Deadline: Oct. 4 AOE adaptive-foundation-models.org

Renjie Liao (@lrjconan) 's Twitter Profile Photo

It means a lot to someone who has worked on Hopfield networks and Boltzmann machines. John and Geoff's foundational work has inspired generations of scientists, and it will keep doing so! I now have a compelling reason to keep teaching and researching these models! Huge Congrats!

Renjie Liao (@lrjconan) 's Twitter Profile Photo

Interested in LLMs for math reasoning? Check out our new work that builds 647K Math QA pairs and A Live Math Benchmark that is kept up-to-date, led by my brilliant students Sadegh Mahdavi Muchen Li and Kaiwen!