#computing #AI #knowledge-representation #reasoning #symbolic-AI #machine-learning #statistical-AI #data-analysis #pattern-recognition #llm Created at 240323 # [Anonymous feedback](https://www.admonymous.co/louis030195) # [[Epistemic status]] #shower-thought Last modified date: 240323 Commit: 0 # Related - [[Readwise/Articles/en.wikipedia.org - Neuro-Symbolic AI - Wikipedia]] - [[Computing/AI thoughts 100323]] - [[Computing/Intelligence/AI from scratch rather than based in human map of the territory]] - [[Computing/Intelligence/Current bottlenecks of artificial intelligence]] - [[Computing/Intelligence/Machine Learning/Semantic search]] # TODO > [!TODO] TODO # Statistical vs symbolic AI Symbolic AI is an approach of developing intelligent systems using symbols or icons that represent meaningful concepts. In symbolic AI, knowledge representation and reasoning are based on the manipulations of symbols and rules. This technique is characterized by a high level of human-like reasoning, so it is sometimes called "rule-based" AI. This approach is used in logical reasoning, machine learning, and expert systems. Statistical AI, also known as machine learning, is based on statistical and probabilistic models that use algorithms to detect patterns in data. It involves using algorithms to statistically analyze patterns and develop models that "learn" from datasets to make predictions or decisions. It uses a large amount of data to recognize patterns and determine relationships between inputs and outputs. Statistical AI is used in diverse areas such as image and speech recognition, natural language processing, and recommendation systems. In summary: - Symbolic AI focuses on knowledge representation and reasoning based on symbols and rules, while statistical AI uses algorithms and data to detect patterns and make predictions. - Symbolic AI is often more human-like in its reasoning, while statistical AI is more data-driven. - Both approaches are used in different areas of AI and have theirown strengths and weaknesses. Symbolic AI can be compared to a bird building a nest using precise and pre-determined behaviors and patterns, while statistical AI can be compared to a bird identifying the location and availability of food sources based on trial-and-error experience and learning from previous successes or failures. Both processes serve different purposes and can be effective in their own ways. In a way, symbolic AI is coming from the emerged representation of knowledge, while statistical AI is coming from the microscopic level of data analysis and pattern recognition. Both approaches have their own advantages and disadvantages, and combining them can lead to more powerful and comprehensive AI systems. However, the challenge lies in developing methods to effectively integrate these different approaches into a cohesive system. <iframe src="https://link.excalidraw.com/readonly/4o4pT1qrIIu9JDZUtunV" width="100%" height="100%" style="border: none;"></iframe>