Table 1.
Term | Definition |
---|---|
Artificial intelligence (AI) |
•Computer systems that perform tasks that normally require human intelligence. For example, visual perception, speech recognition, and decision-making. •Usually involves pattern recognition then followed by an action or a decision |
Machine learning |
•Subdomain of AI •The computer uses algorithms to learn from datasets of past examples to make predictions about new data, as opposed to executing a set of programmed rules. •In classic machine learning, programmers design and tune these algorithms. |
Deep learning |
•Subdomain of machine learning •The computer uses a mathematical structure inspired by neural networks to learn from very large datasets to make predictions about new data. •The neural network builds the algorithms automatically by finding novel relationships between inputs and outputs. •The algorithms cannot be analyzed by humans as they involve 1,000,000 s of small decisions about data. |