SeaPearl: A Constraint Programming Solver guided by Reinforcement\n Learning
Bibliographic record
Abstract
The design of efficient and generic algorithms for solving combinatorial\noptimization problems has been an active field of research for many years.\nStandard exact solving approaches are based on a clever and complete\nenumeration of the solution set. A critical and non-trivial design choice with\nsuch methods is the branching strategy, directing how the search is performed.\nThe last decade has shown an increasing interest in the design of machine\nlearning-based heuristics to solve combinatorial optimization problems. The\ngoal is to leverage knowledge from historical data to solve similar new\ninstances of a problem. Used alone, such heuristics are only able to provide\napproximate solutions efficiently, but cannot prove optimality nor bounds on\ntheir solution. Recent works have shown that reinforcement learning can be\nsuccessfully used for driving the search phase of constraint programming (CP)\nsolvers. However, it has also been shown that this hybridization is challenging\nto build, as standard CP frameworks do not natively include machine learning\nmechanisms, leading to some sources of inefficiencies. This paper presents the\nproof of concept for SeaPearl, a new CP solver implemented in Julia, that\nsupports machine learning routines in order to learn branching decisions using\nreinforcement learning. Support for modeling the learning component is also\nprovided. We illustrate the modeling and solution performance of this new\nsolver on two problems. Although not yet competitive with industrial solvers,\nSeaPearl aims to provide a flexible and open-source framework in order to\nfacilitate future research in the hybridization of constraint programming and\nmachine learning.\n
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 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| 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".