Domain-Independent Game Abstraction using Word Embedding Techniques
The paper presents a domain-independent approach to game abstraction using word embedding techniques. This method allows for the representation of game actions as real-valued vectors, facilitating the abstraction process. Experimental results indicate that while effective, this technique does not surpass specialized algorithms designed for specific games.
- ▪The proposed approach uses word embedding techniques from natural language processing for game abstraction.
- ▪Actions are treated as words, and gameplay data is considered a corpus for training word vectors.
- ▪Experimental results show that the technique is effective but does not outperform specialized algorithms.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Computer Science and Game Theory arXiv:2605.15543 (cs) [Submitted on 15 May 2026] Title:Domain-Independent Game Abstraction using Word Embedding Techniques Authors:Juho Kim, Tuomas Sandholm View a PDF of the paper titled Domain-Independent Game Abstraction using Word Embedding Techniques, by Juho Kim and Tuomas Sandholm View PDF HTML (experimental) Abstract:Many games of interest in the real world are often intractably large, thereby necessitating the use of game abstraction to shrink them in size, typically by many magnitudes.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
