Hello! I'm Logan Stafman, currently a PhD Candidate at Princeton University in the Computer Science department, working with Dr. Michael Freedman. My work currently lies in big data processing as it pertains to approximate computing.
SLAQ: Quality-Driven Scheduling for Machine Learning Applications
Logan Stafman*, Haoyu Zhang*, Andrew Or, Michael J Freedman
Training machine learning models with large datasets can incur significant resource contention on shared clusters. Yet in exploratory settings, better models can be obtained faster by directing resources to jobs with the most potential for improvement. SLAQ is a cluster scheduling system for approximate ML training jobs that aims to maximize the overall job quality. Experiments show that SLAQ achieves a quality improvement of up to 73% and a delay reduction of up to 44%.
Symposium on Cloud Computing '17, Santa Clara [Best Paper] [PDF] [Slides]
Extended Abstract at SysML '18, Stanford, CA
I am extremely enthusiastic about teaching. In the past I have been a TA for the following courses:
In addition to my work, I enjoy juggling, and am an active member of the Princeton Juggling Club. My favorite moment as a juggler is shown below, in which I help fix a rural Vietnamese child's yoyo and show him some tricks.
I also enjoy solving the Rubik's cube at a rather amateur level (PB: 24.55).
I can be contacted at the email below.