Session | ||
TA11 - ML5: Learning algorithms
| ||
Presentations | ||
Online learning via offline greedy algorithms: applications in market design and optimization 1Chicago Booth School of Business, Operations Management; 2MIT Sloan School of Management, Operations Management; 3Google Research Mountain View We study the problem of transforming offline algorithms to their online counterparts, focusing on offline combinatorial problems that are amenable to a constant factor approximation using a greedy algorithm that is robust to local errors. We provide a general offline-to-online framework using Blackwell approachability, producing T^1/2 regret under the full information setting and T^2/3 regret in the bandit setting. We apply our framework to operations problems and produce improved regret bounds. Deep policy iteration with integer programming for inventory management Leonard N Stern School of Business, United States of America In this work, we discuss Programmable Actor Reinforcement Learning (PARL), a policy iteration method that uses techniques from integer programming and sample average approximation. We numerically benchmark the algorithm in complex supply chain settings where optimal solution is intractable and show its performs comparable to, and sometimes better than, state-of-the-art RL and commonly used inventory management benchmarks. Representing random utility choice models with neural networks 1London Business School, United Kingdom; 2INSEAD Motivated by the successes of deep learning, we propose a class of neural network-based discrete choice models, called RUMnets, which is inspired by the random utility maximization (RUM) framework. This model formulates the agents' random utility function using the sample average approximation (SAA) method. We show that RUMnets sharply approximates the class of RUM discrete choice models. We provide analytical and empirical evidence of the predictive power of RUMnets. |