# Intersymbolic AI: Interlinking Symbolic AI and Subsymbolic AI

**Authors:** André Platzer  
**arXiv:** [2406.11563](https://arxiv.org/abs/2406.11563)  
**Date:** 2024-06-17  
**Summary page:** https://www.summarizepaper.com/en/paper/2406.11563v1/

## Key points

- Intersymbolic AI combines symbolic AI (logic, games, planning) with subsymbolic AI
- Authors emphasize the need for integration of both paradigms to enhance AI systems' effectiveness
- Call for collaboration between experts in symbolic and subsymbolic AI
- Intersymbolic AI aims to bridge conscious and subconscious thought in human cognition
- Utilizes symbolic principles to guide subsymbolic learning processes in dynamic systems
- Showcases diverse applications tailored to specific contexts using intersymbolic AI techniques
- Optimism for future advancements in intersymbolic AI and encouragement for further exploration by researchers from diverse backgrounds

## Summary

In this perspective piece, the authors delve into the emerging field of Intersymbolic AI. This approach combines symbolic AI - characterized by building blocks with inherent significance and meaning such as logic, games, and planning - with subsymbolic AI. The authors highlight the complementary strengths and weaknesses of both paradigms and emphasize the need for integration to enhance the effectiveness of AI systems. They call for researchers to explore intersymbolic AI as a way to leverage the benefits of both symbolic and subsymbolic techniques. The key contribution of this work is its call to action for collaboration between experts in symbolic and subsymbolic AI. By drawing parallels to how conscious and subconscious thought enhances human cognition, intersymbolic AI aims to bridge these two worlds in artificial intelligence. It utilizes symbolic principles to guide subsymbolic learning processes in dynamic systems, offering a novel approach to optimizing performance. The authors showcase diverse applications where intersymbolic AI can be applied, highlighting different combinations of symbolic and subsymbolic techniques tailored to specific contexts. This scientific diversity underscores the versatility of intersymbolic AI in addressing challenges across various domains. Looking ahead, the authors express optimism for future advancements in intersymbolic AI and encourage researchers from diverse backgrounds to explore its potential further. By fostering collaboration between experts in both paradigms, there is an opportunity to unlock new insights and innovations that can propel the field forward. In conclusion, this perspective piece underscores the importance of integrating symbolic and subsymbolic approaches in developing more effective AI systems. Through intersymbolic AI, researchers have a unique opportunity to harness the strengths of both paradigms and drive advancements towards more sophisticated problem-solving capabilities.