36 citations · 111 across the 23 of their papers we have counts for
5 papers · 1 filter
An Adiabatic Theorem for Policy Tracking with TD-learning
Neil Walton
We evaluate the ability of temporal difference learning to track the reward function of a policy as it changes over time. Our results apply a new adiabatic theorem that bounds the…
Perturbed Pricing
Neil Walton, Yuqing Zhang
We propose a simple randomized rule for the optimization of prices in revenue management with contextual information. It is known that the certainty equivalent pricing rule, albeit…
Fast Approximate Bayesian Contextual Cold Start Learning (FAB-COST)
Jack R. McKenzie, Peter A. Appleby, Thomas House +1
Cold-start is a notoriously difficult problem which can occur in recommendation systems, and arises when there is insufficient information to draw inferences for users or items. To…
A Short Note on Soft-max and Policy Gradients in Bandits Problems
Neil Walton
This is a short communication on a Lyapunov function argument for softmax in bandit problems. There are a number of excellent papers coming out using differential equations for pol…
Regret Analysis of a Markov Policy Gradient Algorithm for Multi-arm Bandits
Denis Denisov, Neil Walton
We consider a policy gradient algorithm applied to a finite-arm bandit problem with Bernoulli rewards. We allow learning rates to depend on the current state of the algorithm, rath…