Solving the Inferential Frame Problem in the General Game Description Language
Authors: Javier Romero Davila, Abdallah Saffidine, Michael Thielscher
AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experimental results demonstrate that with the help of automatically generated domain knowledge, a significant speedup can thus be obtained for the majority of the game descriptions from the AAAI competition. We present our experimental results and conclude. |
| Researcher Affiliation | Academia | Javier Romero Universit at Potsdam javier@cs.uni-potsdam.de Abdallah Saffidine University of New South Wales abdallahs@cse.unsw.edu.au Michael Thielscher University of New South Wales mit@cse.unsw.edu.au |
| Pseudocode | No | The paper describes logical transformations and rules but does not include any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide explicit access to the source code for the methodology described. It mentions using the 'Flux Player system' which is a third-party tool. |
| Open Datasets | Yes | The benchmark domains have been used in international or local GGP competitions and can be found online.2 http://games.ggp.org |
| Dataset Splits | No | The paper does not specify traditional training, validation, or test dataset splits. It operates on game descriptions and measures performance during minimax search up to a certain depth. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running experiments. |
| Software Dependencies | No | The Flux Player system is based on ECLi PSe Prolog. No version numbers are provided for either software component. |
| Experiment Setup | Yes | Measuring engine speed in GGP is typically done by selecting games from previous GGP competitions and running the game engine related part of standard search algorithms. Following recent work by Schiffel and Bj ornsson (2013), we use the time needed for a naive minimax search to estimate the raw performance. d is the maximal depth reachable in an iterative deepening minimax search within 3 minutes using the classical state transition. We then measure the time (in seconds) needed to perform a full minimax search up to d with the proposed encodings: the original state transition O, the direct translation XD, and the pruned translation XP. |