Causal learning vs. deep learning: on a fatal flaw in machine learning | BBN Times

Photo by Michael Dziedzic found on Unsplash

Data patterns are extracted as quantitative measurements from raw data, structured or unstructured, using standard machine and deep learning methods. This is applicable for both basic algorithms such as logistic regression and more complex algorithms such as neural networks, which may understand deeper patterns from inputs. But there’s a big BUT, and it’s a catastrophic fault in the entire scheme. Machine learning methods, deep learning algorithms, and deep neural networks used in today’s AI can’t detect causality, its components and frameworks, operations and procedures, laws and connections, data and models, and everything else that makes up our world. 

Source: https://www.bbntimes.com/technology/causal-learning-vs-deep-learning-on-a-fatal-flaw-in-machine-learning 

RELATED ARTICLES

Subscribe to our newsletter