Learning through Dialogue Interactions by Asking Questions

Jiwei Li, Alexander H. Miller, Sumit Chopra, Marc'Aurelio Ranzato, Jason Weston
12/15/2016
cs.CLcs.AI

Official code: facebookarchive/MemNN (1,753 stars, last commit 2020-07-28).

Abstract

A good dialogue agent should have the ability to interact with users by both responding to questions and by asking questions, and importantly to learn from both types of interaction. In this work, we explore this direction by designing a simulator and a set of synthetic tasks in the movie domain that allow such interactions between a learner and a teacher. We investigate how a learner can benefit from asking questions in both offline and online reinforcement learning settings, and demonstrate that the learner improves when asking questions. Finally, real experiments with Mechanical Turk validate the approach. Our work represents a first step in developing such end-to-end learned interactive dialogue agents.

View on arXivView PDF

Code Implementations(1)

facebookarchive/MemNNOfficial100%

Memory Networks implementations

1,753368C, LuaSep 16, 20156 years agoNOASSERTION

Similar Papers

A User Simulator for Task-Completion Dialogues

90% match1 repo8032016

Dialogue Learning With Human-In-The-Loop

88% match1 repo1.8k2016

Deep Dyna-Q: Integrating Planning for Task-Completion Dialogue Policy Learning

86% match1 repo1522018

Few-Shot Structured Policy Learning for Multi-Domain and Multi-Task Dialogues

83% match2023

MIMESIS: Learning User Simulators as Training Environments for Interactive Agents

82% match2026

SpeakRL: Synergizing Reasoning, Speaking, and Acting in Language Models with Reinforcement Learning

82% match2025

Cite this paper

@article{li2016learning,
  title  = {Learning through Dialogue Interactions by Asking Questions},
  author = {Jiwei Li and Alexander H. Miller and Sumit Chopra and Marc'Aurelio Ranzato and Jason Weston},
  year   = {2016},
  eprint = {1612.04936},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1612.04936}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1612.04936.svg)](https://codeofpaper.com/paper/1612.04936)

Discussion