The use of AI has been rapidly increasing in the past years, and AI is now everywhere. The amount of digital data available has helped AI techniques evolve into the age of deep learning and large models. For many years, AI has found many applications in cyber security such as intrusion detection and spam detection. More recently, how to make AI systems securer, privacy-friendly and trustworthy has become an important topic.
Do you want to understand how different AI architectures and models work and how they are applied to address cyber security problems and improve the security of systems and computer networks? Do you want to understand how AI models and pipelines can themselves be subject to attacks that compromise their performance, security, privacy, or trustworthiness?
Using cyber security problems and datasets as examples, the module will equip you with the necessary knowledge and hands-on skills about machine learning and deep learning algorithms, models and architectures, how to apply them in real-world applications, and how to interpret results from AI models. In addition, the module also covers the aspects of secure and trustworthy AI in terms of threat models, attacks on AI models and defences against them.
36
Portfolio 1 worth 50% (pass compulsory)
Portfolio 2 worth 50% (pass compulsory)
reassessment: Like for like
On successfully completing the module, students will be able to:
Compare and select the most appropriate AI algorithm among choices provided to address a cyber security problem given a dataset.
Design and implement AI pipelines for cyber security scenarios.
Interpret and compare results from AI models.
Critically evaluate the security and privacy of a given AI pipeline and dataset.
Demonstrate a systematic understanding of possible security or privacy attacks against a given AI pipeline, model, and dataset.
University of Kent makes every effort to ensure that module information is accurate for the relevant academic session and to provide educational services as described. However, courses, services and other matters may be subject to change. Please read our full disclaimer.