AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Dynamic Reward-Based Dueling Deep Dyna-Q: Robust Policy Learning in Noisy Environments

Yangyang Zhao, Zhenyu Wang, Kai Yin, Rui Zhang, Zhenhua Huang, Pei Wang

AAAI 2020 · Reinforcement learning › Visual and representation RL · approach: Scalable RL systems and environments · 33 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
replace DQN with dynamic-reward Dueling DQN inside Dyna-Q (DR-D3Q vs DDQ(20))DDQ(20) · Movie-ticket booking, noise 0.3, epoch 400success rate: 0.3169 -> 0.4120Improved
replace DQN with dynamic-reward Dueling DQN inside Dyna-Q (DR-D3Q vs DDQ(20))DDQ(20) · Movie-ticket booking, noise 0 (noise-free), epoch 300success rate: 0.7080 -> 0.7899Improved
DR-D3Q(20) vs DQN(20) upper bound with real-experience-only extra dataDR-D3Q(20) · Movie-ticket booking, noise 0.3, epoch 400success rate: 0.3725 -> 0.4120Improved
replace dynamic reward with original reward (D3Q, dueling only) (ablation)DR-D3Q · Movie-ticket booking, noise 0.3success rate (learning curve)Hurt
replace Dueling DQN with plain DQN while keeping dynamic reward (DR-DDQ) (ablation)DR-D3Q · Movie-ticket booking, noise 0.3success rate (learning curve)Hurt
human evaluation with 30% erroneous slots: DR-D3Q vs DDQ(20) and DQNDR-D3Q(20) · Human evaluation, noisysuccess rateImproved

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