Evolving dota 2 shadow fiend bots using genetic programming with external memory
Bibliographic record
Abstract
The capacity of genetic programming (GP) to evolve a 'hero' character in the Dota 2 video game is investigated. A reinforcement learning context is assumed in which the only input is a 320-dimensional state vector and performance is expressed in terms of kills and net worth. Minimal assumptions are made to initialize the GP game playing agents - evolution from a tabula rasa starting point - implying that: 1) the instruction set is not task specific; 2) end of game performance feedback reflects quantitive properties a player experiences; 3) no attempt is made to impart game specific knowledge into GP, such as heuristics for improving navigation, minimizing partial observability, improving team work or prioritizing the protection of specific strategically important structures. In short, GP has to actively develop its own strategies for all aspects of the game. We are able to demonstrate competitive play with the built in game opponents assuming 1-on-1 competitions using the 'Shadow Fiend' hero. The single most important contributing factor to this result is the provision of external memory to GP. Without this, the resulting Dota 2 bots are not able to identify strategies that match those of the built-in game bot.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| 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 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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 source (direct Gemma or distilled Codex), 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".