Skip to main content

Trinity College Dublin, The University of Dublin

Menu Search

Module Descriptor School of Computer Science and Statistics

Module CodeST3009
Module NameStatistical Methods for Computer Science
Module Short Title
Semester TaughtSecond Semester
Contact Hours

Lecture: 2 hours per week. Labs: 1 hour per week. Total: 33 hours.

Module PersonnelDoug Leith
Learning Outcomes

When students have completed this module they should be able to:

  • Describe the basic properties of random variables and calculation of probabilities.
  • Explain Bayes theorem and its use in Bayesian inference.
  • Understand confidence intervals and how to calculate them
  • Explain the law of large numbers and understand the importance of the normal distribution.
  • Use linear and logistic regression and apply it to noisy data.
Learning Aims

The module provides an introduction to statistics and probability for computer scientists. The aim is to provide the basic grounding needed for machine learning and algorithm performance analysis.

Module Content

Topics covered in this module include:

  • Experiments, events, probability of an outcome.
  • Conditional probability and Bayes Theorem.
  • Independence.
  • Marginalisation.
  • Mean, variance, covariance
  • Law of Large Numbers, Central Limit Theorem and Normal distribution.
  • Confidence intervals and their calculation using chebyshev/chernoff bounds, central limit theorem, bootstrapping)
  • Maximum likelhood and MAP estimates.
  • Linear regression
  • Logistic Regression
Recommended Reading List

A First Course in Probability, S.Ross.

Module Prerequisites

Basic algebra and programming (we will use Matlab in examples/labs)

Assessment Details

Examination 70%, coursework 30%.  Coursework: 10% weekly assignments and 20% mid-term exam.
Supplemental assessment is by 100% examination.

Academic Year of Data2017/2018