Speakers
All times are in IST
February 13, 2021
9:30 AM - 10:00AM
Dr. Aarthi Ravikrishnan
Postdoctoral Researcher, Genome Institute of Singapore
Title: Gut microbiome recovery post-antibiotic treatment. (Nature Ecology & Evolution)
Human gut houses trillions of diverse microorganisms, and this diversity is perturbed by several factors including the usage of antibiotics. In this talk, I will be presenting our latest metagenome-wide association study that identifies key microbial species involved in gut microbiome recovery post antibiotic treatment.
2. 10:00 AM - 10:30 AM
Anant Shah
Dual Degree Student, Department of Electrical Engineering, IIT Madras
Title: Sequential Ski Rental Problem. (AAMAS 2021)
In this work, multiple instances of the ski rental problem are run using online learning techniques. We propose no regret algorithms to solve the problem
3. 10:30 AM - 11:00 AM
Shiv Kumar Tavker
Dual Degree Student, Department of Mechanical Engineering, IIT Madras
Title: Consistent Plug-In Classifiers for Complex Losses and Constraints. (NeurIPS 2020)
How can we minimize complex loss like F-measure subject to constraints like Fairness or Budget?
4. 11:00 AM - 11:30 AM
Ashutosh Kakadiya
MS Scholar, Department of Computer Science & Engineering, IIT Madras
Title: Relational Boosted Bandits (Rb2). (AAAI 2021)
First Contextual bandit online learning algorithm in relational domain with emphasis on interpretability.
5. 11:30 AM - 12:00 PM
Madhura Pande
MS Scholar, Department of Computer Science & Engineering, IIT Madras
Title: The Heads Hypothesis: A unifying statistical approach towards understanding multi-headed attention in BERT. (AAAI 2021)
A formal, statistical approach of labelling the attention heads of BERT (state-of-the-art NLP model)
6. 12:00 PM - 12:30 PM
Aravind Venugopal
Post Baccalaureate Fellow, RBCDSAI
Title: Reinforcement Learning for Unified Allocation and Patrolling in Signaling Games with Uncertainty. (AAMAS 2021)
Our work focuses on finding solutions for real-world green security problems. Specifically, we present an application of RL to Green Security Games for combating wildlife poaching.
