The Digital Synaptic Neural Substrate: A New Approach to Computational Creativity (1507.07058v2)

Abstract: We introduce a new AI approach called, the 'Digital Synaptic Neural Substrate' (DSNS). It uses selected attributes from objects in various domains (e.g. chess problems, classical music, renowned artworks) and recombines them in such a way as to generate new attributes that can then, in principle, be used to create novel objects of creative value to humans relating to any one of the source domains. This allows some of the burden of creative content generation to be passed from humans to machines. The approach was tested in the domain of chess problem composition. We used it to automatically compose numerous sets of chess problems based on attributes extracted and recombined from chess problems and tournament games by humans, renowned paintings, computer-evolved abstract art, photographs of people, and classical music tracks. The quality of these generated chess problems was then assessed automatically using an existing and experimentally-validated computational chess aesthetics model. They were also assessed by human experts in the domain. The results suggest that attributes collected and recombined from chess and other domains using the DSNS approach can indeed be used to automatically generate chess problems of reasonably high aesthetic quality. In particular, a low quality chess source (i.e. tournament game sequences between weak players) used in combination with actual photographs of people was able to produce three-move chess problems of comparable quality or better to those generated using a high quality chess source (i.e. published compositions by human experts), and more efficiently as well. Why information from a foreign domain can be integrated and functional in this way remains an open question for now. The DSNS approach is, in principle, scalable and applicable to any domain in which objects have attributes that can be represented using real numbers.

• The paper introduces the Digital Synaptic Neural Substrate (DSNS), a technique that recombines numeric attributes from various domains to automate creative generation.
• It applies DSNS in chess problem composition, combining low-quality and unrelated domain data to produce compositions with high aesthetic value.
• Experimental results indicate that integrating diverse domain data can surpass traditional, single-domain AI approaches in creative output quality.

Overview of the Digital Synaptic Neural Substrate Approach

The paper presents a novel artificial intelligence technique termed the Digital Synaptic Neural Substrate (DSNS), designed to enhance computational creativity by integrating attributes from diverse domains such as chess problems, classical music, and artworks. The primary objective of DSNS is to automatically generate creative objects of aesthetic value using a process that mimics the complexity and randomness inherent in human creativity.

Methodology and Approach

DSNS processes involve recombining attributes from different domains to create new, potentially creative objects. Key to this methodology are the attributes from source domains, represented numerically, which are extracted and processed to generate 'deviation values'—these reflect creative differences between objects. The DSNS technique scales by allowing data integration from any domain where objects can be numerically described.

The authors tested DSNS primarily in chess problem composition, leveraging a set of attributes to describe chess problems and other domain objects. The method employed a systematic combination of random and structured attribute selection to produce innovative chess compositions, evaluated through a computational chess aesthetics model.

Experimental Results

Through extensive experiments, the DSNS approach demonstrated its viability. Notably, chess problems generated using low-quality source data combined with unrelated domain data (e.g., photographs of people) achieved comparable or superior quality to those generated from high-quality chess sources. This surprising outcome suggests the potential of cross-domain data integration for enhancing creative output.

Quantitative analyses revealed that DSNS-generated compositions from mixed-domain sources displayed higher aesthetic scores than those derived from single-domain, high-quality sources alone. This highlights the method's potential to transcend boundaries of traditional domain-specific AI techniques.

Implications and Future Directions

The findings contribute significantly to computational creativity, specifically in automating creative processes previously reliant on human intervention. By illustrating that diverse domain data can be effectively amalgamated to enhance creativity in AI, the research opens avenues for further exploration in other creative domains beyond chess.

From a theoretical perspective, the DSNS approach prompts a rethinking of creative processes in AI, proposing that genuine creativity might not require sophisticated domain-specific rules but rather a robust method for diverse data integration.

Looking forward, applications of DSNS could extend to various fields, including music composition, literary creativity, and visual arts, aiding machines in generating creative content that challenges human expectations.

Conclusion

The introduction of the Digital Synaptic Neural Substrate marks a noteworthy advancement in artificial intelligence, providing a scalable, domain-independent framework for creativity. The research underscores the potential for interdisciplinary data blending to revolutionize creative content generation, offering a promising blueprint for future AI developments in creativity and beyond.