AI4C Lecture Series: SimpleMind: How Human-Like Thinking Can Get AI Out of the Lab and Into...

In this presentation, Dr. Matthew Brown of UCLA will discuss how deep neural networks detect patterns in data and have shown versatility and strong performance in many computer vision applications. However, they are susceptible to obvious mistakes that violate common sense concepts and are limited in their ability to use knowledge to guide their decision making. While overall performance metrics may be good, these obvious errors, coupled with a lack of explainability, have prevented widespread adoption for crucial tasks such as medical image analysis. Cognitive AI is broadly defined as enabling human level reasoning and intelligence, and there are emerging approaches that seek to combine neural networks with symbolic reasoning in a best-of-both-worlds scenario. Dr. Brown will introduce SimpleMind, an open-source software environment that supports deep neural networks with knowledge and machine reasoning, improving reliability and trustworthiness and providing explainable decisions. He will demonstrate its benefits in moving AI for medical imaging into patient care.