SDSC8015 - Machine Learning and Control Theory | ||||||||
| ||||||||
| * The offering term is subject to change without prior notice | ||||||||
Course Aims | ||||||||
Machine Learning relies on the theory of optimization. However, the most successful part, which is Deep Learning relies on Control Theory. This is a recent discovery for the Machine Learning community, and it is the object of active research. The deep learning structure is based on a sequence of layers of neural nets. With an infinite number of layers, one obtains a structure amenable to Control Theory. The class will provide all the concepts and methods, in optimization and control theory, which are important and currently used in practice and in research. The models are not simply deterministic. So stochastic control will also be presented. In addition, the connection with the topic of identification of dynamical systems will be explained and developed. Reinforcement learning which is another aspect of Machine Learning, is closely linked with MDP, Markov Decision Processes. We also present Bayesian Learning, with an application in inventory control. | ||||||||
Assessment (Indicative only, please check the detailed course information) | ||||||||
Continuous Assessment: 70% | ||||||||
Examination: 30% | ||||||||
Examination Duration: 3 hours | ||||||||
Detailed Course Information | ||||||||
| SDSC8015.pdf | ||||||||
Useful Links | ||||||||
| School of Data Science | ||||||||
