CS401

MACHINE LEARNING & NEURAL NETWORKS

Credits
5
Year
4
Semester
1
Department
COMPUTER SCIENCE

Overview

Machine learning principles; The probabilistic perspective on machine learning; Supervised, unsupervised, and reinforcement learning; Biological neurons and their relation to linear classifiers; Supervised learning techniques: linear regression, the perceptron learning rule, backpropagation, recurrent networks, deep learning, support vector machines; Unsupervised techniques: k-means clustering, EM of gaussian mixtures, hidden markov models; Reinforcement learning: Policy-valu...

Learning Outcomes

  • Distinguish the different categories of Machine Learning techniques and identify situations in which they might be used
  • Describe the implementation of simple algorithms for machine learning
  • Discuss the adaptation of a standard technique to a given specific problem
  • Describe the implementation and evaluation some standard machine learning algorithms in a general purpose programming language such as Java
  • Obtain a basic understanding of machine learning approaches and neural networks
  • Discuss the relationships between training set, test set, generalisation, cross validation
  • Describe how various machine learning techniques work and what their strengths and limitations are
  • Select and use appropriate machine learning techniques to solve real problems