Comparative Study of Reinforcement Learning in GitHub Pull Request Outcome Predictions
Bibliographic record
Abstract
In the rapidly evolving field of software development, pull-based development models, facilitated by tools such as GitHub, are essential for collaboration. This study explores factors that influence pull request (PR) outcomes and employs two Reinforcement Learning (RL) formalizations, modeled as Markov Decision Processes, for PR outcome prediction. The first model leverages 72 PR features and achieves a G-mean score of 0.82664, while the second focuses solely on PR discussions, resulting in a G-mean of 0.88372. Using a specially designed reward function, these RL formalizations strategically address data imbalance and excel in mimicking both single-stage and multi-stage PR review processes. They outperform baseline models (Random Forest, X G Boost, and a Naive Bayes baseline) across various data splits-namely 80/20, 50/50, and 20/80-and are particularly effective at predicting PR rejections. The study also makes its datasets publicly available for future research.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".