AI Engines in Short-video Era: Eating 100 Trillion Parameters

Description
Speaker
This event is hosted by SF Big Analytics Group. https://www.meetup.com/SF-Big-Analytics

The ‘short-video’ has been proven to be one of the most successful product forms in the 4G era, witnessed by Tiktok, Instagram, Snapchat, Kuaishou, etc. AI has demonstrated its huge potentials to improve traditional businesses and create new product features in `short-video’ apps, e.g., beautifying filter, live stream, recommendation, Ads, games, risk control, etc. The common issue in the `short-video’ era faced by AI applications is the high volume of data (imagine that each user generates a behavior for every 15-20 seconds) and the huge size of the model (imagine that a recommendation model may have dozens of trillion parameters). Therefore, the acceleration of AI applications in the ‘short-video’ era greatly depend on the efficiency of the process from model training to inference and online serving.

This speaker, rely on his experiences at Kuaishou, will mainly introduce two *Pytorch* based AI engines to accelerate the training process in two major scenarios: Bagua for the “dense” scenario (e.g., CV/NLP/speech) and Persia for the sparse scenario (recommendation/Ads/search). To pursue the extreme efficiency, both engines conduct joint optimization and design of algorithms and system implementation. Bagua improves the popular open source tools such as Horovod by 30% efficiency; Persia is 7 times faster than other open sourced engines such as XDL by Alibaba, and is capable of scaling up to 100 trillion recommendation model parameters.

Ji Liu

Dr. Ji Liu received his Ph.D in computer science and his bachelor degree in automation from University of Wisconsin-Madison and University of Science and Technology of China, respectively. He founded the AI platform department, Seattle AI lab, FeDa Ads intelligence lab, and the game AI lab at Kwai Inc. His team supports 2000+ scientists and engineers’ daily development from model training to inference and online serving
  • Date: Jan 11, 17:00 (US Pacific Time)
  • Fee: Free
  • Available Seats: 429 (max 500)
  • Help? Send Question
Watch Recording