A Deep Reinforcement Learning Chatbot

Iulan Serban (julianserban <at> gmail (point) com)

MILA, DIRO

Le 11 octobre 2017 à 11 h 30

Salle 3195, Pavillon André-Aisenstadt


We present MILABOT: a deep reinforcement learning chatbot developed by the Montreal Institute for Learning Algorithms (MILA) for the Amazon Alexa Prize competition. MILABOT is capable of conversing with humans on popular small talk topics through both speech and text. The system consists of an ensemble of natural language generation and retrieval models, including sequence-to-sequence and latent variable neural network models. By applying reinforcement learning to crowd-sourced data and real-world user interactions, the system has been trained to select an appropriate response from the models in its ensemble. The system has been evaluated through A/B testing with real-world users, where it was found to perform excellent.


Pour recevoir les annonces hebdomadaires par courriel, envoyez un message à l'adresse majordomo@iro.umontreal.ca

Liste des autres séminaires