Hi I'm Giulio.
I'm a third year PhD Student in the Computer Science Department at Carnegie Mellon University advised by Dave Andersen. I am generally interested in the intersection of systems and machine learning. Recently, I've been thinking about how systems can benefit from ML -- for instance, can large-scale pre-training learn general representations that generalize well to tasks in different parts of the systems stack? Are there classes of problems where ML can fill the (potentially) huge gulf between the best hand-engineered heuristic and the optimal oracular policy?
Last summer, I was an intern at Google Brain with Martin Maas worked on learning shared representations for storage system tasks from distributed tracing. I am excited to be continuing this collaboration this semester as a Student Researcher at Google.
In undergrad, I worked on probabilistic programming languages with Stuart Russell and ML model serving with Joey Gonzalez. Previously, I also worked in a nanocrystal synthesis lab and at Google as a software engineer.
I have also had the privilege to TA CS 61B (Data Structures and Algorithms) under Josh Hug and CS61BL under Sarah Kim and Alan Yao. I also TA'd for CS 189 (Introduction to Machine Learning) under Ben Recht and Jitendra Malik. This semester, I'll be TA'ing 10-701, CMU's PhD-level Introduction to Machine Learning course.
Giulio Zhou, Martin Maas. Multi-Task Learning for Storage Systems. Machine Learning for Systems Workshop, NeurIPS 2019.
Angela H. Jiang, Daniel L.-K. Wong, Giulio Zhou, David G. Andersen, Jeffrey Dean, Gregory R. Ganger, Gauri Joshi, Michael Kaminksy, Michael Kozuch, Zachary C. Lipton, Padmanabhan Pillai. Accelerating Deep Learning by Focusing on the Biggest Losers. arxiv preprint arXiv:1910.00762
Christopher Canel, Thomas Kim, Giulio Zhou, Conglong Li, Hyeontaek Lim, David G. Andersen, Michael Kaminsky, Subramanya R. Dulloor. Scaling Video Analytics on Constrained Edge Nodes. SysML, 2019.
Giulio Zhou, Subramanya Dulloor, David G. Andersen, Michael Kaminsky. EDF: Ensemble, Distill, and Fuse for Easy Video Labeling. arXiv preprint arXiv:1812.03626
Daniel Crankshaw, Xin Wang, Giulio Zhou, Michael J. Franklin, Joseph E. Gonzalez, Ion Stoica. Clipper: A Low-Latency Online Prediction Serving System. NSDI, 2017.