English

Uncertainty Estimates for Efficient Neural Network-based Dialogue Policy Optimisation

Machine Learning 2017-12-04 v1 Computation and Language Machine Learning Neural and Evolutionary Computing

Abstract

In statistical dialogue management, the dialogue manager learns a policy that maps a belief state to an action for the system to perform. Efficient exploration is key to successful policy optimisation. Current deep reinforcement learning methods are very promising but rely on epsilon-greedy exploration, thus subjecting the user to a random choice of action during learning. Alternative approaches such as Gaussian Process SARSA (GPSARSA) estimate uncertainties and are sample efficient, leading to better user experience, but on the expense of a greater computational complexity. This paper examines approaches to extract uncertainty estimates from deep Q-networks (DQN) in the context of dialogue management. We perform an extensive benchmark of deep Bayesian methods to extract uncertainty estimates, namely Bayes-By-Backprop, dropout, its concrete variation, bootstrapped ensemble and alpha-divergences, combining it with DQN algorithm.

Keywords

Cite

@article{arxiv.1711.11486,
  title  = {Uncertainty Estimates for Efficient Neural Network-based Dialogue Policy Optimisation},
  author = {Christopher Tegho and Paweł Budzianowski and Milica Gašić},
  journal= {arXiv preprint arXiv:1711.11486},
  year   = {2017}
}

Comments

Accepted at the Bayesian Deep Learning Workshop, 31st Conference on Neural Information Processing Systems (NIPS 2017)

R2 v1 2026-06-22T23:02:36.590Z