Deep Reinforcement Learning

Neural Approximate Dynamic Programming for On-Demand Ride-Pooling

We add future information to ride-pooling assignments by using a novel extension to Approximate Dynamic Programming.

Solving Online Threat Screening Games using Constrained Action Space Reinforcement Learning

We reformulate Threat Screening Games, a kind of Stackelberg Security Game, as an MDP with constraints on the action space.

My Ride-pooling Journey

A blog post about the joyride my paper 'Neural Approximate Dynamic Programming for On-Demand Ride-Pooling' took me on.