Will S. Monroe

TL;DR: This work simulates dialogues between two virtual agents, using policy gradient methods to reward sequences that display three useful conversational properties: informativity, non-repetitive turns, coherence, and ease of answering.

TL;DR: This paper proposed using adversarial training for open-domain dialogue generation, where the generator is trained to generate sequences that are indistinguishable from human-generated dialogue utterances, and the outputs from the discriminator are used as rewards for the generator.

TL;DR: This work applies adversarial training to open-domain dialogue generation, training a system to produce sequences that are indistinguishable from human-generated dialogue utterances, and investigates models for adversarial evaluation that uses success in fooling an adversary as a dialogue evaluation metric, while avoiding a number of potential pitfalls.

TL;DR: This paper proposes a general methodology to analyze and interpret decisions from a neural model by observing the effects on the model of erasing various parts of the representation, such as input word-vector dimensions, intermediate hidden units, or input words.

TL;DR: The authors apply deep reinforcement learning to model future reward in chatbot dialogue, using policy gradient methods to reward sequences that display three useful conversational properties: informativity (nonrepetitive turns), coherence, and ease of answering.