With the explosion of data and increase in demand for understanding data using machine learning techniques, traditional machine learning solutions are embracing cloud based solutions to meet large scale data and computing needs. This course attempts to augment the present theoretical treatment of machine learning models in the context of advanced cloud based implementations.
Introduction: principles of cloud computing systems, cloud architectures, service models;
Cloud architectures and service platform: cloud architecture and infrastructure design, dynamic deployment, various cloud service examples;
Introduction to Machine Learning: supervised learning methods, unsupervised learning methods, introduction to deep learning models.
Life cycle: data storage, data cleaning, data transformation, model training, model testing, model deployment and integration, model monitoring and feedback;
Cloud programming with Spark: understanding resilent distributed dataset (RDD), Spark MLlib;
TensorFlow: Declaring tensors, place holders and variables, working with matrices, declaring operations, data sources, implementing loss functions, batch training.
1. K.Hwang, Cloud Computing for Machine Learning and Cognitive Applications, MIT Press, 2017.
2. R.O.Duda, P.E.Hart and D.G.Strok, Pattern Classification, 2nd Edn., John Wiley & Sons Pvt. Ltd, 2001.
3. M.R.Karim and M.M.Kaysar, Large Scale Machine Learning with Spark, Packt Publishing, 2016.
4. N.McClure, TensorFlow Machine Learning Cookbook, Packt Publishing Limited, 2017.