Designing AI systems for deep learning recommendations and more
Speaker: Carole-Jean Wu

Abstract:
Over the past decade, the amount of computing power for AI has increased 300,000 times. The latest natural language processing model is fed with more than trillion parameters, while the memory requirements of neural recommendation and ranking models have grown from hundreds of gigabytes to terabytes. This talk introduces the underinvested deep learning personalization and recommendation systems to the overall research community. Training advanced personalization and recommendation models at industrial scale requires the largest number of compute cycles of any deep learning use case on Facebook. For AI inference, recommendation use cases require even higher compute cycles of 80%. What are the key system challenges facing industrial-scale neural personalization and recommendation models? This talk will highlight recent advances in AI systems development for deep learning recommendations and the implications for infrastructure optimization capabilities within the machine learning systems stack. Systems research for deep learning recommendations and AI in general is at a nascent stage. This lecture concludes with research directions for building and designing responsible AI systems – that are fair, efficient and environmentally sustainable.

Speaker biography:
Carole-Jean Wu is a technical lead and manager at Facebook AI Research – SysML. Her work is in the area of ​​computer systems architecture, with a particular emphasis on energy and memory efficient systems. Her research has focused on designing systems for implementing machine learning at scale, such as for personalized recommendation systems and mobile deployment. Overall, she is interested in addressing systems challenges to enable efficient, responsible AI execution. Carole-Jean is chair of the MLPerf Recommendation Benchmark Advisory Board, co-chair of MLPerf Inference, and serves as a director on the MLCommons Board. Carole-Jean received her MA and Ph.D. from Princeton and B.Sc. from Cornell. She is the recipient of the NSF CAREER Award, Facebook AI Infrastructure Mentorship Award, the IEEE Young Engineer of the Year Award, the Science Foundation Arizona Bisgrove Early Career Scholarship, and the Intel PhD Fellowship, in addition to a number of Best Paper awards.

—
0:00 Starts soon
4:46 Presentation
42:05 Discussion

The Stanford MLSys Seminar is hosted by Dan Fu, Karan Goel, Fiodar Kazhamiaka and Piero Molino, Chris Ré and Matei Zaharia.

Twitter:
https://twitter.com/realDanFu​
https://twitter.com/krandiash​
https://twitter.com/w4nderlus7

—

Check our website for the class schedule: http://mlsys.stanford.edu
Join our mailing list to receive weekly updates: https://groups.google.com/forum/#!forum/stanford-mlsys-seminars/join

#machinelearning #ai #artificialintelligence #systems #mlsys #computerscience #stanford #fair #facebookai #recommendersystems #deeplearning

Please take the opportunity to connect and share this video with your friends and family if you find it useful.