Neural symbolic processing combines two important types of artificial intelligence: neural networks and symbolic logic.
While neural networks recognize patterns in data, symbolic logic works with clear rules and symbols, similar to a language or mathematics.
The combination of these two approaches helps computers to both learn flexibly and draw complex, explainable conclusions.
This makes it possible to structure knowledge while simultaneously learning from examples.
In this way, intelligent systems can better understand what they are doing and are easier to control.