# User:Arvindprakash

**About me :**

I am a final year bachelors student, doing Information Systems at BITS Pilani, Goa Campus. I am currently doing a research internship at IIT Madras as part of my final year project.

**Mentors :**

Dr. Balaraman Ravindran

Dr. Nandan Sudarsanam

**Areas of interest :**

Reinforcement learning, Recommender Systems

**Working on :**

**Linear bandits and applications - **Bandit problems deal with situations where an agent needs to make a choice out of given possible choices (called arms) over several iterations. The goal of the problem is to maximize the total reward over all iterations. Contextual bandits are an extension of the normal bandit problem, wherein an agent observes a set of features, or a 'context', before making a choice. Linear bandits are a specialized case of contextual bandits, where the rewards observed are assumed to be linearly dependant on the corresponding context. I am currently working on linear bandits and their applications on real world datasets.