Reinforcement Learning for Stochastic Networks, Toulouse

Session

Parallel session: Some applications of reinforcement learning to networks

Jun 19, 2024, 1:30 PM

A001 (ENSEEIHT)

Description

Organizer and chair: Vivek Borkar

53. Learning LP-indices in Average-Reward Restless Multi-Armed Bandits

Dr Konstantin Avrachenkov (INRIA Sophia Antipolis)

6/19/24, 1:30 PM

Restless Multi-Armed Bandits (RMABs) are extensively used in scheduling,
resource allocation, marketing and clinical trials, just to name a few
application areas. RMABs are Markov Decision Processes with two actions
(active and passive modes) for each arm and with a constraint on the
number of active arms per time slot. Since in general RMABs are
PSPACE-complete, several heuristics such...

90. Optimal congestion signaling under customer heterogeneity with private types

Prakirt Jhunjhunwala (Columbia Business School)

6/19/24, 2:00 PM

In an unobservable queue, where customers lack the complete wait time information, a throughput-maximizing server aims to exploit the information asymmetry by strategically signaling coarse congestion information to incentivize customers' arrival into the system. The customers make a calculated decision about joining the queue by creating a belief of their utility given the congestion signal...

45. Energy Management in Smart Grids Using Finite Horizon SOR Q-learning

Shalabh Bhatnagar (Indian Institute of Science)

6/19/24, 2:30 PM

The smart grid is comprised of
different microgrids and is supported by different technologies. The microgrid has to make a lot of decisions, and this has to be automated for efficiency. We have Markov Decision Processes as a frame-
work for sequential decision-making under uncertainty and
reinforcement learning techniques to learn the optimal decisions. In this work, we formulate this...