Definition of AI, ML, and DL
Subscribe to receive short posts and illustrations about explainable AI.
Before diving into explainable AI, it will help to understand the definitions of AI, ML, and DL.
Artificial Intelligence (AI): Set of systems that imitate human cognitive abilities.
Machine Learning (ML): Subset of AI where systems are programmed to perform tasks without having specific sets of rules. ML systems are trained using data and learn from experience. High-quality training data is necessary to create a high-quality model.
Deep Learning (DL): Subset of ML where systems are learning hidden patterns and important features by themselves through repetitive training and adaptation. Deep neural networks (DNN) use a structure similar to the human neural system to analyze different factors.
Examples: AI, ML, and DL Predicting Disease Diagnosis
Artificial Intelligence: A human programmed a rule-based system (a decision tree) that analyzes patient data and results in a prediction for whether the patient will be diagnosed.
Machine Learning: A human programmed a system using previous data and indicating important factors such as patient age, family history, and lifestyle factors to consider for the system to predict whether a patient will be diagnosed.
Deep Learning: A human programmed a system using previous data only (with no indication of which factors should be considered by the system) to predict whether a patient will be diagnosed. An example is Deep Patient!
Want to dig deeper into this topic?
To learn about some of the many applications of AI, visit AI applications: Top 10 Artificial Intelligence Applications, part of Simplilearn's AI for Beginners tutorial.
For a more detailed, technical explanation AI/ML/DL, check out Artificial Intelligence vs. Machine Learning vs. Deep Learning: Essentials from Serokell's blog, which does a great job illustrating each!
Next, check out the Introduction to Explainable AI!