Draft:Feature Space Search
Submission declined on 14 June 2024 by Encoded (talk). This submission is not adequately supported by reliable sources. Reliable sources are required so that information can be verified. If you need help with referencing, please see Referencing for beginners and Citing sources.
Where to get help
How to improve a draft
You can also browse Wikipedia:Featured articles and Wikipedia:Good articles to find examples of Wikipedia's best writing on topics similar to your proposed article. Improving your odds of a speedy review To improve your odds of a faster review, tag your draft with relevant WikiProject tags using the button below. This will let reviewers know a new draft has been submitted in their area of interest. For instance, if you wrote about a female astronomer, you would want to add the Biography, Astronomy, and Women scientists tags. Editor resources
| ![]() |
Submission declined on 24 March 2024 by KylieTastic (talk). This draft's references do not show that the subject qualifies for a Wikipedia article. In summary, the draft needs multiple published sources that are: Declined by KylieTastic 5 months ago.
| ![]() |
Submission declined on 17 March 2024 by KylieTastic (talk). This draft's references do not show that the subject qualifies for a Wikipedia article. In summary, the draft needs multiple published sources that are: Declined by KylieTastic 5 months ago.
| ![]() |
Feature Space Search (FESS) is a novel single-agent search algorithm designed to address limitations of traditional search methods in complex domains. Unlike other search algorithms that use features and combine them into a single value, FESS treats the features as an abstract space (feature space) to be explored. It guides the search process in the domain space through exploration within this feature space, without relying on an evaluation or heuristic function.
Generality
[edit]FESS is designed as a general approach, offering a potentially more efficient solution for various problems by exploring features independently and providing multi-objective guidance to the search process. This makes it well-suited for challenging problems where finding a solution is a priority, even if not the most optimal one. While initially applied to Sokoban puzzles, its flexibility allows for adaptation to other domains with complex problem-solving requirements.
Development
[edit]Motivated by the need for better search strategies, the algorithm was presented at the 2020 IEEE Conference on Games (CoG) by Yaron Shoham and Prof. Jonathan Schaeffer.[1]
Performance
[edit]Empirical evaluations have shown strong performance compared to other search methods in Sokoban puzzles.[2]
FESS Process
[edit]FESS works by discovering new cells in feature-space until a path is established between the start cell and the target cell. This is done by iterating over the cells and attempting to find unvisited neighbours. Moves are played in the domain space with the aim of discovering new feature-space cells.
External links
[edit]References
[edit]- ^ Yaron Shoham; Jonathan Shaeffer (2020). The FESS Algorithm: A Feature Based Approach to Single-Agent Search (PDF). 2020 IEEE Conference on Games (CoG). Osaka, Japan: IEEE. doi:10.1109/CoG47356.2020.9231929.
- ^ Christopher Olson; Lars Wagner; Alexander Dockhorn (2023). Evolutionary Optimization of Baba Is You Agents (PDF). 2023 IEEE Congress on Evolutionary Computation (CEC). Chicago, USA: IEEE. p. 3. doi:10.1109/CEC53210.2023.10253977.