Research map / Reinforcement learning / Offline RL
RL-Duet: Online Music Accompaniment Generation Using Deep Reinforcement Learning
Nan Jiang, Sheng Jin, Zhiyao Duan, Changshui Zhang
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 |
|---|---|---|---|
| Train the generation policy with RL-Duet rather than MLE | Online accompaniment generation · Subjective paired comparison | votes preferring generated music: 58.42% | Improved |
| Use RL-Duet rather than MLE | Online accompaniment generation · Bach Chorale test set | PCH earth moving distance: 0.0067 -> 0.0057 | Improved |
| Use RL-Duet rather than MLE | Online accompaniment generation · Bach Chorale test set | NLH earth moving distance: 0.039 -> 0.042 | Mixed |
| Use the multi-hold pitch representation instead of single-hold (ablation) | RL-Duet generation model · Music generation | diversity | Improved |
| Use the multi-hold pitch representation instead of single-hold (ablation) | Reward models (b), (c), and (d) · Music generation | rhythmic naturalness | Hurt |
| Replace the transplanted SequenceTutor rules with RL-Duet's learned reward approach | Online accompaniment generation · Subjective user study | generation quality | Improved |
Later papers that compared with this method
1 later accepted paper reports RL-Duet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- When Counterpoint Meets Chinese Folk Melodies (NeurIPS 2020 · better on 4 of 4 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Adaptive Accompaniment with ReaLchords (ICML 2024 · similarity 0.90)
- MusicRL: Aligning Music Generation to Human Preferences (ICML 2024 · similarity 0.87)
- Duolando: Follower GPT with Off-Policy Reinforcement Learning for Dance Accompaniment (ICLR 2024 · similarity 0.85)
- JEN-1 Composer: A Unified Framework for High-Fidelity Multi-Track Music Generation (AAAI 2025 · similarity 0.85)
- Compose with Me: Collaborative Music Inpainter for Symbolic Music Infilling (AAAI 2025 · similarity 0.84)
- CMI-RewardBench: Evaluating Music Reward Models with Compositional Multimodal Instruction (ICML 2026 · similarity 0.84)
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.
