Search Results for author: Pawel Budzianowski

Found 4 papers, 1 papers with code

Variational Cross-domain Natural Language Generation for Spoken Dialogue Systems

1 code implementation • WS 2018 • Bo-Hsiang Tseng, Florian Kreyssig, Pawel Budzianowski, Inigo Casanueva, Yen-chen Wu, Stefan Ultes, Milica Gasic

Cross-domain natural language generation (NLG) is still a difficult task within spoken dialogue modelling.

Diversity Sentence +2

6

Paper
Code

Nearly Zero-Shot Learning for Semantic Decoding in Spoken Dialogue Systems

no code implementations • 14 Jun 2018 • Lina M. Rojas-Barahona, Stefan Ultes, Pawel Budzianowski, Iñigo Casanueva, Milica Gasic, Bo-Hsiang Tseng, Steve Young

This paper presents two ways of dealing with scarce data in semantic decoding using N-Best speech recognition hypotheses.

speech-recognition Speech Recognition +2

Paper
Add Code

Neural User Simulation for Corpus-based Policy Optimisation for Spoken Dialogue Systems

no code implementations • 17 May 2018 • Florian Kreyssig, Inigo Casanueva, Pawel Budzianowski, Milica Gasic

The ABUS is based on hand-crafted rules and its output is in semantic form.

Diversity Reinforcement Learning +3

Paper
Add Code

Sample-efficient Actor-Critic Reinforcement Learning with Supervised Data for Dialogue Management

no code implementations • WS 2017 • Pei-Hao Su, Pawel Budzianowski, Stefan Ultes, Milica Gasic, Steve Young

Firstly, to speed up the learning process, two sample-efficient neural networks algorithms: trust region actor-critic with experience replay (TRACER) and episodic natural actor-critic with experience replay (eNACER) are presented.

Deep Reinforcement Learning Dialogue Management +3

Paper
Add Code

Cannot find the paper you are looking for? You can Submit a new open access paper.