Research map / Reinforcement learning / Imitation learning
Generative Adversarial Regularized Mutual Information Policy Gradient Framework for Automatic Diagnosis
Yuan Xia, Jingbo Zhou, Zhenhui Shi, Chao Lu, Haifeng Huang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add mutual information reward to vanilla policy gradient (ablation) | policy gradient · Dxy | diagnosis accuracy: 0.731 -> 0.749 | Improved |
| add mutual information reward to vanilla policy gradient (ablation) | policy gradient · Dxy | average turns: 3.11 -> 2.76 | Improved |
| add GAN discriminator reward to vanilla policy gradient (ablation) | policy gradient · Dxy | diagnosis accuracy: 0.731 -> 0.758 | Improved |
| add GAN discriminator reward to vanilla policy gradient (ablation) | policy gradient · Dxy | average turns: 3.11 -> 3.4 | Hurt |
| combine MI and GAN rewards versus MI only (ablation) | PG + MI · Dxy | diagnosis accuracy: 0.749 -> 0.769 | Improved |
| combine MI and GAN rewards versus GAN only (ablation) | PG + GAN · Dxy | diagnosis accuracy: 0.758 -> 0.769 | Improved |
| combine MI and GAN rewards versus GAN only (ablation) | PG + GAN · Dxy | average turns: 3.4 -> 2.68 | Improved |
| policy gradient with fixed reward over MLE pretrained LSTM (ablation) | pretrained LSTM · Dxy | diagnosis accuracy: 0.643 -> 0.731 | Improved |
| add MI and GAN rewards to vanilla policy gradient (ablation) | policy gradient · Dxy | match rate: 0.212 -> 0.179 | Hurt |
Closest papers by meaning
- REFUEL: Exploring Sparse Features in Deep Reinforcement Learning for Fast Disease Diagnosis (NeurIPS 2018 · similarity 0.84)
- SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient (AAAI 2017 · similarity 0.84)
- Deep Reinforcement Learning for Cost-Effective Medical Diagnosis (ICLR 2023 · similarity 0.84)
- MTDiag: An Effective Multi-Task Framework for Automatic Diagnosis (AAAI 2023 · similarity 0.83)
- Diaformer: Automatic Diagnosis via Symptoms Sequence Generation (AAAI 2022 · similarity 0.83)
- Improving Reward Model Generalization from Adversarial Process Enhanced Preferences (ICML 2025 · similarity 0.82)
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.
