Artificial Intelligence Concepts: Applied Statistics (Online)

Unit 1: Introduction to Probability Theory

• Randomness versus vagueness
• Frequentist and Bayesian interpretations of probability
• Sample spaces and the set-theoretical model of probability
• The axioms of probability
• Conditional probability

Unit 2: Bayesian Learning

• Biased coins: limitations of the frequentist approach
• Advantages of the Bayesian approach for Machine Learning
• Bayesian estimation: maximising the posterior distribution
• Optimal Bayes, Gibbs and Naive Bayes classifiers

Unit 3: Bayesian Networks

• Independent probabilities, pairwise independence and mutual independence
• Conditional independence
• Graph theory and notation
• Constructing Bayesian networks
• Bayesian networks for prediction

Unit 4: Introduction to Statistics

• Random variables, expected values and standard deviations
• Sample means and the update formula
• Computing point estimates
• Confidence intervals, Chebyshev’s theorem and the empirical rule
• Performing hypothesis tests
• Visualising data: histograms, box plots and scatter plots

Unit 5: Self-Organising Maps

• The concept of competitive learning
• Self-organising maps and their applications
• Training self-organising maps
• Variations on self-organising maps
• Visual representations of self-organising maps

Unit 6: Information Theory

• Binary and multi-class classification problems
• Binary decision rules for binary classification problem
• Entropy and mutual information
• Decision trees for binary classification problems
• Constructing efficient decision trees via mutual information

Unit 7: Ensemble Methods

• Applications and suitability of ensemble methods
• Bootstrapping
• Boosting to improve model accuracy
• Naive Bayes and ensemble methods

Unit 8: Regression

• Linear models with dependent and independent variables
• Regression versus classification
• Calculating regression coefficients in a linear regression

Unit 9: Working with Data

• Big data analytics: challenges and opportunities
• Describing the properties of datasets
• Data analysis pipelines
• Big data and big corporations

Unit 10: Risks and Benefits of Big Data

• Exploiting big data analysis for prediction, adaptability and value delivery
• Concrete examples of successful implementations
• The legal and technical challenges of big data applications
• Internal and external security concerns regarding big data
• Biases in big data analytics

There is no essential reading associated with this course.

To earn credit (CATS points) for your course you will need to register and pay an additional £10 fee for each course you enrol on. You can do this by ticking the relevant box at the bottom of the enrolment form or when enrolling online. If you do not register when you enrol, you have up until the course start date to register and pay the £10 fee.

See more information on CATS point

Coursework is an integral part of all online courses and everyone enrolled will be expected to do coursework, but only those who have registered for credit will be awarded CATS points for completing work at the required standard. If you are enrolled on the Certificate of Higher Education you need to indicate this on the enrolment form but there is no additional registration fee.

Assignments are not graded but are marked either pass or fail.

All students who successfully complete this course, whether registered for credit or not, are eligible for a Certificate of Completion. Completion consists of submitting the final course assignment. Certificates will be available, online, for those who qualify after the course finishes.

• To introduce fundamental concepts in Artificial Intelligence from Probability and Statistics
• To study a wide range of Artificial Intelligence approaches arising from Applied Statistics
• To identify the possibilities and pitfalls presented by Big Data and Artificial Intelligence

By the end of this course, students should:

• Understand fundamental concepts in Artificial Intelligence from Probability and Statistics
• Understand how Artificial Intelligence draws on statistical foundations
• Have detailed knowledge of specific Applied Statistics methods
• Be able to identify suitable Applied Statistics approaches for Artificial Intelligence applications
• Be able to critically assess the potential impact and pitfalls of Big Data for Artificial Intelligence.

You will be set two pieces of work for the course. The first of 500 words is due halfway through your course. This does not count towards your final outcome but preparing for it, and the feedback you are given, will help you prepare for your assessed piece of work of 1,500 words due at the end of the course. The assessed work is marked pass or fail.

English Language Requirements

We do not insist that applicants hold an English language certification, but warn that they may be at a disadvantage if their language skills are not of a comparable level to those qualifications listed on our website. If you are confident in your proficiency, please feel free to enrol. For more information regarding English language requirements please follow this link: https://www.conted.ox.ac.uk/about/english-language-requirements

FHEQ level 4, 10 weeks, approx 10 hours per week, therefore a total of about 100 study hours.