JamesLaudon

Introducing a novel mixture-of-experts routing algorithm, called Expert Choice, that can achieve optimal load balancing between experts while allowing heterogeneous token-to-expert mapping. Learn how it’s done at goo.gle/3OdpO9t

👏Congratulations to our Professor @anaklimovic for receiving the prestigious #FacebookResearchAward! The award aims to support innovative solutions to the challenges that still remain in the #datamanagement community.bit.ly/3uiTisz @ETH_en @SystemsGroupETH

Excited to share that our work has been published in Nature! Our RL agent generates chip layouts in just a few hours, whereas human experts can take months. These superhuman AI-generated layouts were used in Google's latest AI accelerator (TPU-v5)! nature.com/articles/s4158…

Thrilled to announce that our work on RL for chip floorplanning was published in Nature & used in production to design next generation Google TPUs, with potential to save thousands of hours of engineering effort for each next generation ASIC: nature.com/articles/s4158… (1/7)

Check out new work on ML-driven design and exploration of custom accelerators, showing how #MachineLearning facilitates architecture exploration by rapidly identifying high-performing configurations across a range of applications. Learn more ↓ goo.gle/3rngiEh

On behalf of the entire Google Research & @GoogleAI communities, I'm excited to share an overview of some of our research in 2020. Thanks to everyone who helped make this work possible! ai.googleblog.com/2021/01/google…