Oct 6, 2023 : CSE Bytes
Prof. B. Ravindran
Title: Reinforcement Learning with Structured Actions and Policies
Abstract: Deep Reinforcement Learning has been very successful in solving a variety of hard problems. But many RL architectures treat the action as coming from an unordered set or from a bounded interval. It is often the case that the actions and policies have a non-trivial structure that can be exploited for more efficient learning. In this talk, I will present several scenarios in which taking advantage of the structure leads to more efficient learning.
Prof. Kartik Nagar
Title: Safe programming for distributed, replicated systems.
Abstract: Modern cloud-based software services often replicate data across multiple geographically distributed locations to tolerate partial failures of servers and to minimise latency by bringing data closer to the user. This has introduced a completely new paradigm of programming called replicated data types, which embraces a programming model with multiple copies of data which can be independently updated with lazy synchronisation to ensure consistency. In this talk, we will introduce one instance of this paradigm called mergeable replicated data types, and describe some of our efforts in verifying the correctness of such implementations.
Event Venue - CS15
Timing - 3:30pm
Date - 6th October, 2023
