Kumar Krishna Agrawal

graduate student at eecs, berkeley. previously research@google, math/cs at iit kharagpur

research: simple, efficient algorithms for

opportunities: if you are at berkeley and interested in ml/systems research; please reach out!

about / teaching / writing / github / gscholar


[blog] compute efficient model selection for self-supervised learning

[lecture] guest lecture on CV in Practice at CS182 deep learning, uc berkeley

Prior Knowledge-Guided Attention in Self-Supervised Vision Transformers, 2022
Kevin Miao, Kumar Krishna Agrawal, Akash Gokul, Suzanne Petryk, Raghav Singh, Adam Yala, Joseph Gonzalez, Kurt Keutzer, Trevor Darrell, Colorado J Reed
arXiv / code

selected publications

Neural Population Geometry across model scale: A tool for cross-species functional comparison of visual brain regions
Arna Ghosh, Kumar Krishna Agrawal, Zahraa Chorghay, Arnab Kumar Mondal, Blake Richards
Computational and Systems Neuroscience (COSYNE) 2023

Assessing representation quality in SSL by measuring eigenspectrum decay
Kumar Krishna Agrawal*, Arna Ghosh*, Arnab Kumar Mondal*, Blake Richards
Neural Information Processing Systems (NeurIPS), 2022
openReview / code / blog

Octopus : Low-Latency & Adaptive Perception Pipelines
Gur-Eyal Sela, Ionel Gog, Justin Wong, Kumar Krishna Agrawal, Sukrit Kalra, Peter Schafhalter, Xiangxi Mo, Xin Wang, Bharath Balaji, Ion Stoica, Joseph Gonzalez
European Conference in Computer Vision (ECCV), 2022
arXiv / code

Learning from an Exploring Demonstrator: Optimal Reward Estimation for Bandits
Wenshuo Guo, Kumar Krishna Agrawal, Aditya Grover, Vidya Muthukumar, Ashwin Pananjady
International Conference on Artificial Intelligence and Statistics (AISTATS), 2022
Workshop on Theory of Reinforcement Learning, ICML 2021
Workshop on Human-AI Collaboration in Sequential Decision-Making , ICML 2021 (spotlight)

Discrete Flows: Invertible Generative Models for Discrete Data
Dustin Tran, Keyon Vafa, Kumar Krishna Agrawal, Laurent Dinh, Ben Poole
Neural Information Processing Systems (NeurIPS), 2019

GANSynth: Adversarial Neural Audio Synthesis
Jesse Engel, Kumar Krishna Agrawal, Shuo Chen, Ishaan Gulrajani, Chris Donahue, Adam Roberts
International Conference on Learning Representations, (ICLR) 2019
arXiv / Magenta blog / samples

Discriminator Actor Critic: Addressing sample inefficiency and reward bias in Adversarial Imitation Learning
Ilya Kostrikov, Kumar Krishna Agrawal, Debidatta Dwibedi, Sergey Levine, Jonathan Tompson
International Conference on Learning Representations (ICLR) 2019

Towards Mixed Optimization for Reinforcement Learning with Program Synthesis
Surya Bhupatiraju*, Kumar Krishna Agrawal*, Rishabh Singh
Workshop on Neural Abstract Machines and Program Induction, ICML 2018


Designing, Visualizing and Understanding Deep Neural Networks
fall 2022 @ ucberkeley

Introduction to Machine Learning
spring 2022, spring 2021 @ ucberkeley

Foundations of Machine Learning to Interact with a Dynamic World
fall 2021 @ ucberkeley

Depth First Learning : Learning to Understand Machine Learning
fellowship / whitepaper

(Template credits!)