AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

BBQ-Networks: Efficient Exploration in Deep Reinforcement Learning for Task-Oriented Dialogue Systems

Zachary Lipton, Xiujun Li, Jianfeng Gao, Lihong Li, Faisal Ahmed, Li Deng

AAAI 2018 · Reinforcement learning › Policy gradient and actor-critic · approach: Regularized scalable deep value learning · 122 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use Bayesian Q-network Thompson sampling instead of DQN explorationDialogue policy · Full DomainSuccess Rate: 0.2693 -> 0.5031Improved
Use Bayesian Q-network Thompson sampling instead of DQN explorationDialogue policy · Domain ExtensionSuccess Rate: 0.3503 -> 0.7120Improved
Add VIME intrinsic reward to BBQN-MAP (ablation)BBQN-MAP · Full DomainSuccess Rate: 0.5031 -> 0.4856Hurt
Add VIME intrinsic reward to BBQN-MAP (ablation)BBQN-MAP · Domain ExtensionSuccess Rate: 0.6852 -> 0.6813Mixed
Use Monte Carlo sampled target values instead of MAP targets (ablation)BBQN · Full DomainSuccess Rate: 0.5031 -> 0.4877Hurt
Use Monte Carlo sampled target values instead of MAP targets (ablation)BBQN · Domain ExtensionSuccess Rate: 0.6852 -> 0.6722Mixed
Combine VIME intrinsic reward with BBQN-MC (ablation)BBQN-MC · Domain ExtensionSuccess Rate: 0.6722 -> 0.7120Improved
Use uniform random exploration in a hybrid with Thompson sampling (ablation)BBQN · Movie-booking taskperformanceNo clear effect
Pre-fill replay buffer with successful rule-based dialogues (ablation)BBQN and DQN · Movie-booking tasksuccessImproved
Increase replay-buffer spiking from 100 to 1000 dialogues (ablation)BBQN-MAP · Movie-booking tasklong-term performanceHurt
Train without replay-buffer spiking (ablation)BBQN and DQN · Movie-booking tasksuccessHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy