Research challenges in deep reinforcement learning-based join query optimization
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The order in which relations are joined and the physical join operators used are two aspects of query plans which have a significant impact on the execution latency of join queries. However, the set of valid query plans grows exponentially with the number of relations to be joined. Hence, it becomes computationally expensive to enumerate all such plans for a complex join query. Recently, several deep reinforcement learning (DRL) based approaches propose using neural networks to construct a query plan. They demonstrate that efficient query plans can be found without exhaustively enumerating the search space. We integrated our implementation of a DRL-based solution to optimize join order and operators into the PostgreSQL query optimizer. In practice, we found limitations in the quality of the query plans chosen which are not addressed in existing approaches. In this paper we highlight some of these limitations and propose future research challenges along with potential solutions.
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.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 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 it