Machine Learning

From BarikWiki
Jump to: navigation, search

Machine Learning

Introduction

  • Any computer program that improves its performance at some task through experience. That is "A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improved with experience E."
  • Final design can usually be split into four models: Performance System, Critic, Generalizer, and Experiment Generator.
  • Choices in designing a learning program: Determine Type of Training Experience, Determine Target Function, Determine Representation of Learned Function, Determine Learning Algorithm.

Concept Learning

  • Concept Learning. Inferring a boolean-valued function from training examples of its input and output.
  • The set of items over which the concept is defined is called the set of instances (X). The concept or function to be learned is called the target concept (c) so that c: X -> {0, 1}. D is sometimes the entire training set.
  • Informally, the inductive learning hypothesis. Any hypothesis found to approximate the target function well over a sufficiently large set of training examples will also approximate the target function well over other unobserved examples.