New! Jan 2019: We're presenting a case for a GPU multiplexing

New! Dec 2018: Our paper Revec: Program Rejuvenation through Revectorization was accepted to Compiler Construction 2019!

New! November 2018: Our paper Dynamic Space-Time Scheduling for GPU Inference was accepted to LearningSys (at NeurIPS 2018)!

November 2018: DeepScale's research team released DSCnet: Replicating Lidar Point Clouds with Deep Sensor Cloning

Intro

I’m a Ph.D. student at UC Berkeley and a Research Scientist at DeepScale. As a Ph.D. student, I am advised by Ion Stoica and Joseph Gonzalez. I am a part of the RISE Lab, BAIR Lab and Berkeley Deep Drive Lab.

Recent projects


The case for GPU multitenancy
Paper link

Revec: Program Rejuvenation through Revectorization
Paper link // Github [TBA]
Accepted paper, Intl. Conference on Compiler Construction 2019

Dynamic Space-Time Scheduling for GPU Inference
Paper link // Github
Accepted to LearningSys'18 at NeurIPS 2018
We demonstrate 2.5x to 4.5x speedups for GPU inference for deep learning workloads through a novel GPU multitenancy approach.

DSCnet: Replicating Lidar Point Clouds with Deep Sensor Cloning
Paper link
Latest paper from the DeepScale research team. We present a technique to replicate a $\$$70k LiDAR sensor with inexpensive sensors.

Scalable Architecture for Anomaly Detection and Visualization in Power Generating Assets
Paper link // Slides
Accepted to HPBDC'17 at IPDPS 2017
We present a scalable system for time-series anomaly detection that horizontally scales to sensor networks with >400k QPS.

Spotting Suspicious Reviews via (Quasi-)clique Extraction
Paper link
Accepted poster at IEEE Security and Privacy 2015
We uncover suspicious activity by well-organized reviewer rings sponsored by Yelp. Our work sheds light on Yelp's little-known paid review operations.


Research publications