Lessons from Contextual Bandit Learning in a Customer Support Bot
Abstract
In this work, we describe practical lessons we have learned from successfully using contextual bandits (CBs) to improve key business metrics of the Microsoft Virtual Agent for customer support. While our current use cases focus on single step reinforcement learning (RL) and mostly in the domain of natural language processing and information retrieval we believe many of our findings are generally applicable. Through this article, we highlight certain issues that RL practitioners may encounter in similar types of applications as well as offer practical solutions to these challenges.
Cite
Text
Karampatziakis et al. "Lessons from Contextual Bandit Learning in a Customer Support Bot." ICML 2019 Workshops: RL4RealLife, 2019.Markdown
[Karampatziakis et al. "Lessons from Contextual Bandit Learning in a Customer Support Bot." ICML 2019 Workshops: RL4RealLife, 2019.](https://mlanthology.org/icmlw/2019/karampatziakis2019icmlw-lessons/)BibTeX
@inproceedings{karampatziakis2019icmlw-lessons,
title = {{Lessons from Contextual Bandit Learning in a Customer Support Bot}},
author = {Karampatziakis, Nikos and Kochman, Sebastian and Huang, Jade and Mineiro, Paul and Osborne, Kathy and Chen, Weizhu},
booktitle = {ICML 2019 Workshops: RL4RealLife},
year = {2019},
url = {https://mlanthology.org/icmlw/2019/karampatziakis2019icmlw-lessons/}
}