SORTS: INTEGRATING SOAR WITH A REAL-TIME STRATEGY GAME Investigators
Bibliographic record
Abstract
The goal of this project is to interface Soar to a real-time strategy (RTS) game. RTS games, such as StarCraft, WarCraft, and Command and Conquer, are multi-player strategy games where a player needs to handle many different tasks- planning base layouts, managing economies, organizing attacks, responding to enemy attacks, and even diplomacy. This is all occuring in real time (as the name implies), presenting an extremely rich, challenging environment for an AI system. Making this environment accessible to Soar provides many opportunities for both utilizing existing capabilities and the development of new capabilities. 1.1 ORTS Overview The Open Real Time Strategy software is a highly configurable game engine used to play real time strategy (RTS) games [2]. The main purpose for ORTS is to serve as an open source, open interface RTS game engine for RTS AI tournaments. ORTS is undergoing active development as of July 2006 at the University of Alberta under the direction of Michael Buro. There are several reasons why ORTS is especially suitable for use in AI tournaments. It has a (relatively) straightforward C++ API, making interfacing with your favorite AI system easy. All the specific game mechanics, ranging from types of units, actions, and physics, are specified via C++ style scripts called blueprints. This means that ORTS can be easily configured to simulate a wide range of environments, from arbitrarily simple ones like Wumpus World to complex ones like Starcraft. Finally, ORTS has a client/server architecture in which the server maintains the state of the world and only report to the clients information they are supposed to have for a fair game. This is in contrast to most commercial RTS games, in which each client maintains the entire world state and prevents the player from accessing forbidden information such as other players ’ locations only by hiding them from the GUI. The result is that ORTS is impervious to ”memory hack ” cheats that are widespread in commercial RTS games. This feature is particularly important if tournaments are to be run across the Internet.
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 | 0.001 | 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 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".