AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

When Counterpoint Meets Chinese Folk Melodies

NeurIPS 2020 · Reinforcement learning › RL from feedback · approach: GRPO stability and variants · 5 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 FolkDuet (RL with style and interaction rewards) instead of MLE baselinecountermelody generator · Chinese folk melodies test setPCH (earth mover distance): 0.017 -> 0.008Helped
use FolkDuet instead of MLE baselinecountermelody generator · Chinese folk melodies test setNLH (earth mover distance): 0.036 -> 0.014Helped
use FolkDuet instead of MLE baselinecountermelody generator · Chinese folk melodies test setpitch interval PI (dataset 2.73): 3.02 -> 2.44Helped
use FolkDuet instead of MLE baselinecountermelody generator · Chinese folk melodies test setinter-onset interval IOI (dataset 2.36): 2.87 -> 2.16Helped
use FolkDuet instead of MLE baselinecountermelody generator · Chinese folk melodies test setpitch count per bar (dataset 3.90): 4.21 -> 3.96Helped
use FolkDuet instead of MLE baselinecountermelody generator · Chinese folk melodies test setkey consistency: 0.78 -> 0.85Helped
use FolkDuet instead of MLE baselinecountermelody generator · Chinese folk melodies test setinteraction reward: -0.30 -> 0.13Helped
use RL-Duet (Bach-style RL, no style reward) instead of MLEcountermelody generator · Chinese folk melodies test setpitch interval PI (dataset 2.73): 3.02 -> 4.02Hurt
use RL-Duet instead of MLEcountermelody generator · Chinese folk melodies test setkey consistency: 0.78 -> 0.71Hurt
use RL-Duet instead of MLEcountermelody generator · Chinese folk melodies test setNLH (earth mover distance): 0.036 -> 0.055Hurt
use FolkDuet instead of MLE baselinecountermelody generator · Listening test, 160 participantslistening test harmonic, melodic appealingness and folk style (0-4 scale)Helped
evaluate on participants with musical training vs withoutMLE baseline · Listening test, 160 participantslistening test scoresMixed

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