|
Arjun Devraj Email: adevraj [at] cs [dot] cornell [dot] edu I'm a third-year PhD student in Computer Science at Cornell University, where I'm advised by Rachee Singh. My research interests are broadly in systems and networking, with a focus on machine learning systems. Recently, I have been working on improving communication bottlenecks to speed up distributed machine learning workloads. I am also interested in traffic engineering for wide-area networks. In my research, I enjoy combining algorithmic techniques with insights from hardware, such as optical interconnects. My work is supported by the NSF Graduate Research Fellowship. Before graduate school, I spent two years as a software engineer at Meta, where I worked on privacy infrastructure for the ads recommender system. Previously, I received my bachelor's degree in Computer Science, summa cum laude, from Princeton University, where I worked with Jennifer Rexford on network privacy and with Tom Griffiths and Qiong Zhang on computational cognitive science. Email / Google Scholar / LinkedIn / GitHub |
|
Publications |
|
Efficient AllReduce with Stragglers
Arjun Devraj, Eric Ding, Abhishek Vijaya Kumar, Robert Kleinberg, Rachee Singh
|
|
Stable and Fault-Tolerant Decentralized Traffic Engineering
Arjun Devraj, Umesh Krishnaswamy, Ying Zhang, Karuna Grewal, Justin Hsu, Éva Tardos, Rachee Singh
|
|
Reconfigurable Torus Fabrics for Multi-tenant ML
Abhishek Vijaya Kumar, Eric Ding, Arjun Devraj, Darius Bunandar, Rachee Singh
|
|
In-Network Analog AllReduce for ML with Programmable Integrated Photonics
Arjun Devraj, Bill Owens, Daniel Pérez-López, Rachee Singh
|
|
HEDGE: Traffic Engineering with Probabilistic Link Capacities
Arjun Devraj, Bill Owens, Umesh Krishnaswamy, Ying Zhang, Rachee Singh
|
|
The Reality of Chasing Shannon's Limit in Optical Wide-Area Networks
Arjun Devraj, Bill Owens, Rachee Singh
|
|
Chip-to-chip photonic connectivity in multi-accelerator servers for ML
Abhishek Vijaya Kumar, Arjun Devraj, Darius Bunandar, Rachee Singh
|
|
A case for server-scale photonic connectivity
Abhishek Vijaya Kumar, Arjun Devraj, Darius Bunandar, Rachee Singh
|
|
Reconciling categorization and memory via environmental statistics
Arjun Devraj, Thomas L. Griffiths, Qiong Zhang
|
|
REDACT: Refraction Networking from the Data Center
Arjun Devraj, Liang Wang, Jennifer Rexford
|
|
The Dynamics of Exemplar and Prototype Representations Depend on Environmental Statistics
Arjun Devraj, Qiong Zhang, Thomas L. Griffiths
|
Talks and Presentations |
Efficient AllReduce with Stragglers
|
Stable and Fault-Tolerant Decentralized WAN Traffic Engineering
|
HEDGE: Traffic Engineering with Probabilistic Link Capacities
|
The Reality of Chasing Shannon's Limit in Optical WANs
|
|
Source code adapted from Jon Barron's website. |