Evolution for automatic assessment of the difficulty of sokoban boards
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
Many games have a collection of boards with the difficulty of an instance of the game determined by the starting configuration of the board. Correctly rating the difficulty of the boards is somewhat haphazard and required either a remarkable level of understanding of the game or a good deal of play-testing. In this study we explore evolutionary algorithms as a tool to automatically grade the difficulty of boards for a version of the game sokoban. Mean time-to-solution by an evolutionary algorithm and number of failures to solve a board are used as a surrogate for the difficulty of a board. Initial testing with a simple string-based representation, giving a sequence of moves for the sokoban agent, provided very little signal; it usually failed. Two other representations, based on a reactive linear genetic programming structure called an ISAc list, generated useful hardness-classification information for both hardness surrogates. These two representations differ in that one uses a randomly initialized population of ISAc lists while the other initializes populations with competent agents pre-trained on random collections of sokoban boards. The study encompasses four hardness surrogates: probability-of-failure and mean time-to-solution for each of these two representations. All four are found to generate similar information about board hardness, but probability-of-failure with pre-evolved agents is found to be faster to compute and to have a clearer meaning than the other three board-hardness surrogates.
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 | 0.000 | 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 it