An Ensemble Model with Ranking for Social Dialogue
This work addresses the challenge of creating engaging AI dialogue systems for real-world applications, though it is incremental as it builds on existing ensemble and ranking methods.
The authors tackled the problem of open-domain social dialogue by developing an ensemble system for the Amazon Alexa Prize, which achieved coherent and engaging conversations with real users for up to 20 minutes.
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.