CGSC 2002 Lecture Notes - Lecture 16: Confusion Matrix, False Positives And False Negatives, Support Vector Machine
19 Mar 2018
School
Department
Course
Professor
Document Summary
John anderson proposes the following arguments in favor of production rules: Arguments for the psychological reality of production rules include: Evidence of the appropriateness of rules in describing many aspects of skilled behavior. Evidence of the ability to predict the details of that behavior under a production-rule description. Learning is a key aspect of intelligence . Thus, many ai methods focus on machine learning (ml). Learning involves the ability to generalize from experience. Memorization, on the other hand, is trivial and often useless, especially for a computer. Machine learning involves construction and analysis of models that can learn from data to make predictions on that data. During the learning phase, the model is presented with sample inputs and the corresponding outputs (the data is labelled). The goal is to learn how to map inputs to outputs. Output labels are not provided during the learning phase, leaving it on its own to find structure in the input.
