Canonical Orderings on Grids
Authors: Nathan R. Sturtevant, Steve Rabin
IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The goal of our experimental results is to measure the performance of our new search algorithms, CA*, BJPS, and CD, as well as the weighted counterparts of CA* and BJPS. We will evaluate these algorithms by looking at the total time to solve a problem (in ms), the number of expansions, the number of node generations (this includes states that were analyzed during successor generation but not added to open), and the number of states on open at the end of the search. |
| Researcher Affiliation | Academia | Nathan R. Sturtevant Department of Computer Science University of Denver Denver, CO, USA sturtevant@cs.du.edu Steve Rabin Department of Computer Science Digi Pen Institute of Technology Redmond, WA, USA steve.rabin@gmail.com |
| Pseudocode | Yes | Algorithm 1 Best-First Search BFS(start, goal() 1: Add start to open 2: while open not empty do 3: Get best from open 4: if best is goal then 5: Return path 6: end if 7: for all successors s of best do 8: if s in open then 9: Update cost to s in open 10: else if s not in closed then 11: Add s to open 12: end if 13: end for 14: Add best to closed 15: end while |
| Open Source Code | No | The paper mentions that their implementations are based on existing A* and custom JPS code, but does not provide any links or explicit statements about making their code publicly available. |
| Open Datasets | Yes | Our experiments are run on the Dragon Age: Origins maps from the Moving AI repository [Sturtevant, 2012]. |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, or detailed splitting methodology) for training, validation, and testing. |
| Hardware Specification | Yes | All of our experiments were run on a 2.3 GHz Intel Core i7 laptop with 16 GB of RAM. |
| Software Dependencies | No | The paper describes software components like |
| Experiment Setup | Yes | Our priority queue is implemented with a heap, with grid-based open and closed lists. Our implementation of CD and CA* are based upon existing A* code. Our implementation of BJPS is based on a custom JPS implementation. All searches use the octile distance heuristic and ties are broken towards larger g-costs. |