Sudeep Raja is a Doctoral student in the IEOR Department at Columbia University, advised by Prof. Shipra Agrawal. His research interests are Sequential decision-making under uncertainty, specifically online convex optimization, multi-armed bandits and reinforcement learning of structured MDPs. He aims to design efficient algorithms with provable guarantees for online decision-making. Sudeep holds a Master of Science in Computer Science from the University of Massachusetts Amherst and a Bachelor of Technology in Computer Science and Engineering from the Indian Institute of Technology Kharagpur.
My CV: CV
Email: sudeepraja.putta at columbia.edu
Simons Institute Profile: 1, 2
News
- I will be a Visiting Graduate Student in the Data Driven Decision Processes program at the Simons Institute in Fall 2022.
- I am attending ALT 2022 in Paris, where I will present my work on Scale Free Adversarial Multi Armed Bandits.
- I will be a Visiting Graduate Student in the Theory of Reinforcement Learning program at the Simons Institute in Fall 2020.
- I will be attending COLT’19 and STOC’19 at Phoenix.
- I will be attending the MIFODS workshop on Non-convex optimization and deep learning in January.
- My paper, Exponential Weights on the Hypercube in Polynomial Time, has been accepted at AISTATS 2019. This is joint work with Abhishek Shetty. I will be presenting this work at Naha, Okinawa, Japan
- I received the Sudha Mishra and Rajesh Jha Scholarship.
- In summer 2018 I will be interning with Navin Goyal at MSR Bangalore.
Publications
2022
- Scale Free Adversarial Multi Armed Bandits
Sudeep Raja Putta, Shipra Agrawal
ALT 2022.
2019
- Exponential Weights on the Hypercube in Polynomial Time
Sudeep Raja Putta, Abhishek Shetty
AISTATS 2019.
2018
- Exponential Weights on the Hypercube in Polynomial Time
Sudeep Raja Putta
EWRL 2018. (This paper is superseded by the AISTATS 2019 paper)
2017
- Pure Exploration in Episodic Fixed-Horizon Markov Decision Processes
Sudeep Raja Putta, Theja Tulabandhula
AAMAS 2017. Short paper. - Efficient Reinforcement Learning via Initial Pure Exploration
Sudeep Raja Putta, Theja Tulabandhula
RLDM 2017.
Quotes
Don’t worry about the overall importance of the problem; work on it if it looks interesting. I think there’s a sufficient correlation between interest and importance. — David Blackwell