BBQ-Networks: Efficient Exploration in Deep Reinforcement Learning for Task-Oriented Dialogue Systems
Artificial Intelligence
2017-11-21 v2 Computation and Language
Machine Learning
Abstract
We present a new algorithm that significantly improves the efficiency of exploration for deep Q-learning agents in dialogue systems. Our agents explore via Thompson sampling, drawing Monte Carlo samples from a Bayes-by-Backprop neural network. Our algorithm learns much faster than common exploration strategies such as \epsilon-greedy, Boltzmann, bootstrapping, and intrinsic-reward-based ones. Additionally, we show that spiking the replay buffer with experiences from just a few successful episodes can make Q-learning feasible when it might otherwise fail.
Cite
@article{arxiv.1711.05715,
title = {BBQ-Networks: Efficient Exploration in Deep Reinforcement Learning for Task-Oriented Dialogue Systems},
author = {Zachary Lipton and Xiujun Li and Jianfeng Gao and Lihong Li and Faisal Ahmed and Li Deng},
journal= {arXiv preprint arXiv:1711.05715},
year = {2017}
}
Comments
Duplicate of article already in the arXiv: arXiv:1608.05081