An ensemble model with ranking for social dialogue

Papaioannou, Ioannis; Curry, Amanda Cercas; Part, Jose L; Shalyminov, Igor; Xu, Xinnuo; Yu, Yanchao; Dušek, Ondřej; Rieser, Verena; Lemon, Oliver

Abstract

Open-domain social dialogue is one of the long-standing goals of Artificial Intelligence. This year, the Amazon Alexa Prize challenge was announced for the first time, where real customers get to rate systems developed by leading universities worldwide. The aim of the challenge is to converse “coherently and engagingly with humans on popular topics for 20 minutes”. We describe our Alexa Prize system (called ‘Alana’) consisting of an ensemble of bots, combining rule-based and machine learning systems, and using a contextual ranking mechanism to choose a system response. The ranker was trained on real user feedback received during the competition, where we address the problem of how to train on the noisy and sparse feedback obtained during the competition.

Citation

Papaioannou, I., Curry, A. C., Part, J. L., Shalyminov, I., Xu, X., Yu, Y., …Lemon, O. (2017, December). An ensemble model with ranking for social dialogue. Paper presented at NIPS 2017 Conversational AI Workshop, Long Beach, US

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: NIPS 2017 Conversational AI Workshop
• Start Date: Dec 8, 2017
• Deposit Date: Jun 28, 2023
• Related Public URLs: https://doi.org/10.48550/arXiv.1712.07558