The Tip of the Iceberg: How to make 面向系统的ML work

摘要

Machine Learning has become a powerful tool to improve computer 系统 and there is a significant amount of research ongoing both in academia and industry to tackle 系统 problems using Machine Learning. Most work focuses on learning patterns and replacing heuristics with these learned patterns to solve 系统 problems such as compiler optimization, query optimization, failure detection, indexing, caching. 然而, solutions that truly improve 系统 need to maintain the efficiency, availability, reliability, and maintainability of 系统 while integrating Machine Learning into 该系统. In this talk, I will cover the key aspects and surprising joys of designing, implementing and deploying 面向系统的ML solutions based on my experiences of building and deploying these 系统 at Google.

生物

Deniz is a Software Engineer in the 面向系统的ML team at Google Brain Research. She received her PhD in distributed 系统 from Cornell University in 2017 under the supervision of Robbert van Renesse, specializing on consensus protocols and self-adapting 系统. Since her PhD she has worked on large-scale database 系统 and learned 系统. Currently, she is focusing on improving the state-of-the-art in 系统 using Machine Learning, mainly focusing on caching and indexing. Her work has appeared in top-tier conferences and workshops such as SOSP and 面向系统的ML at NeurIPS.

时间和地点

October 25, 2022 @ 1:30PM in MH 225