200 citations · 1.1k across the 39 of their papers we have counts for
7 papers · 1 filter
Ethical Challenges in Data-Driven Dialogue Systems
Peter Henderson, Koustuv Sinha, Nicolas Angelard-Gontier +4
The use of dialogue systems as a medium for human-machine interaction is an increasingly prevalent paradigm. A growing number of dialogue systems use conversation strategies that a…
Z-Forcing: Training Stochastic Recurrent Networks
Anirudh Goyal, Alessandro Sordoni, Marc-Alexandre Côté +2
Many efforts have been devoted to training generative latent variable models with autoregressive decoders, such as recurrent neural networks (RNN). Stochastic recurrent models have…
ACtuAL: Actor-Critic Under Adversarial Learning
Anirudh Goyal, Nan Rosemary Ke, Alex Lamb +4
Generative Adversarial Networks (GANs) are a powerful framework for deep generative modeling. Posed as a two-player minimax problem, GANs are typically trained end-to-end on real-v…
Sparse Attentive Backtracking: Long-Range Credit Assignment in Recurrent Networks
Nan Rosemary Ke, Anirudh Goyal, Olexa Bilaniuk +4
A major drawback of backpropagation through time (BPTT) is the difficulty of learning long-term dependencies, coming from having to propagate credit information backwards through e…
Variational Walkback: Learning a Transition Operator as a Stochastic Recurrent Net
Anirudh Goyal, Nan Rosemary Ke, Surya Ganguli +1
We propose a novel method to directly learn a stochastic transition operator whose repeated application provides generated samples. Traditional undirected graphical models approach…
A Deep Reinforcement Learning Chatbot
Iulian V. Serban, Chinnadhurai Sankar, Mathieu Germain +15
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 capa…
Twin Networks: Matching the Future for Sequence Generation
Dmitriy Serdyuk, Nan Rosemary Ke, Alessandro Sordoni +3
We propose a simple technique for encouraging generative RNNs to plan ahead. We train a "backward" recurrent network to generate a given sequence in reverse order, and we encourage…