Author pages are created from data sourced from our academic publisher partnerships and public sources.
Share This Author
Logan: A Distributed Online Log Parser
- Amey Agrawal, Rohit Karlupia, Rajat Gupta
- Computer ScienceIEEE 35th International Conference on Data…
- 8 April 2019
A data-driven log parser is trained on the authors' new Apache Spark dataset, the largest application log dataset yet, and a distributed online algorithm is implemented to accommodate for the large volume of data.
Delog: A Privacy Preserving Log Filtering Framework for Online Compute Platforms
- Amey Agrawal, Abhishek Dixit, Darshil Kapadia, Rohit Karlupia, Vikram Agrawal, Rajat Gupta
- Computer ScienceArXiv
- 13 February 2019
A privacy preserving framework which can be employed by Platform as a Service (PaaS) providers to utilize the user logs generated on the platform while protecting the potentially sensitive logged data and a distributed log parsing algorithm which leverages Locality Sensitive Hashing.
Singularity: Planet-Scale, Preemptive and Elastic Scheduling of AI Workloads
This work presents Singularity, Microsoft’s globally distributed scheduling service for highly-efficient and reliable execution of deep learning training and inference workloads, and shows that the resulting efficiency and reliability gains are achieved with negligible impact on the steady-state performance.
Delog: A High-Performance Privacy Preserving Log Filtering Framework
- Amey Agrawal, Abhishek Dixit, Rohit Karlupia
- Computer ScienceIEEE International Conference on Big Data (Big…
- 1 December 2019
A privacy-preserving framework that can be employed by Platform as a Service (PaaS) providers to utilize the user logs generated on the platform while protecting the potentially sensitive logged data is described.
Learning Digital Circuits: A Journey Through Weight Invariant Self-Pruning Neural Networks
This work uses the existing framework of binarized networks to find performant topologies by constraining the weights to be either, zero or one, and shows that such topologies achieve performance similar to standard networks while pruning more than 99% weights.