Keynote Speakers

Mari Ostendorf

Professor, Department of Electrical Engineering,
University of Washington, USA

Title: Understanding the User in Socialbot Conversations

Abstract: Much past research on human-computer dialog has addressed task-oriented scenarios, but there is growing interest in building systems with social interaction capabilities, from companionship chitchat to information and opinion exchange. For systems that emphasize social interaction (e.g. a socialbot), user modeling can be especially important -- people have different tastes in conversation topics as well as different interaction styles. This talk looks at the user in spoken interactions enabled by Sounding Board, a socialbot developed for the 2017 Amazon Alexa Prize competition, which enabled collection of millions of conversations with real users. We describe mechanisms for characterizing user variation and first steps towards predicting conversational preferences.

Ingrid Zukerman

Professor of Computer Science
Faculty of Information Technology
Monash University, Australia

Title: Interpreting spoken referring expressions in physical settings

Abstract: Human discourse is sometimes ambiguous or inaccurate, and may contain words that are unknown to a dialogue system. These phenomena are exacerbated in a spoken dialogue system due to ASR error. I describe our approach to address these problems when interpreting spoken referring expressions in physical settings, and outline a recent experiment in the generation of responses to these referring expressions.

Milica Gasic

Lecturer, Department of Engineering
University of Cambridge, UK

Title: Towards natural conversation with machines using deep learning

Abstract: Deep learning has made a revolution in machine learning, natural language processing and computer vision. In this talk, I will explain how deep learning can help solve some of the problems that dialogue modelling is facing. These include: scalable belief tracking, policy optimisation for large action spaces, and data-driven user modelling. I will also briefly advertise an initiative of the Cambridge Dialogue Systems Group to address the problem of evaluation of dialogue systems.

Very regrettably, the keynote by Prof. Stede was cancelled due to his sickness

Manfred Stede

Professor of Applied Computational Linguistics in the Dept. of Linguistics,
Research Cluster Cognitive Science, University of Potsdam, Germany

Title: Automated arguing: From mining to synthesizing

Abstract: Argumentation mining, the detection of arguments and their structure in text, has enjoyed increasing interest in recent years. We outline the goals of the field and sketch the current state of the art for detecting components of arguments and building structural representations. For the dialogue community, however, the reverse task of generating argumentative text is of similar (if not equal) importance. This step has so far received much less attention, though. We briefly look back at some early work on text generation for evaluative and argumentative genres, and then report on some current experiments on synthesizing "new" text from "old" argument components.