Post date: Nov 10, 2015 3:22:22 PM
MBA Course Notes to
MACHINE LEARNING
From robotics, speech recognition, and analytics to finance and social network analysis, machine learning has become one of the most useful set of scientific tools of our age. With this course we want to bring interested students and researchers from a wide array of disciplines up to speed on the power and wide applicability of machine learning. The ultimate aim of the course is to equip you with all the modelling and optimization tools you’ll need in order to formulate and solve problems of interest in a machine learning framework. We hope to help build these skills through lectures and reading materials which introduce machine learning in the context of its many applications, as well as by describing in a detailed but user-friendly manner the modern techniques from nonlinear optimization used to solve them.
Bài giảng Khóa học Thạc sỹ
HỌC MÁY
Từ robot, nhận dạng giọng nói, phân tích tài chính và phân tích mạng xã hội, học máy đã trở thành một trong những công cụ khoa học hữu ích nhất hiện nay. Với khóa học này, chúng tôi muốn mang lại cho sinh viên và các nhà nghiên cứu quan tâm trong các ngành khác nhau khả năng khai thác sức mạnh và ứng dụng rộng rãi của máy học. Mục đích cuối cùng của khóa học là trang bị cho bạn với tất cả các công cụ mô hình hóa và tối ưu hóa cần thiết để xây dựng và giải quyết các vấn đề quan tâm trong máy học. Chúng tôihy vọng sẽ giúp xây dựng các kỹ năng thông qua các bài giảng và tài liệu đọc giới thiệu máy học cùng vớinhiều ứng dụng của nó, cũng như bằng cách mô tả một cách chi tiết và dễ hiểu, các kỹ thuật hiện đại từ tối ưu hóa phi tuyến để giải quyết bài toán.
Data acquisition devices, computers and networks are generating data to an unprecedented level that postulates the development and utilization of powerful, robust and adaptive learning solutions in order to accomplish various challenging tasks such as pattern recognition, time series modeling, optimization, decision support, diagnosis, text mining, and multimedia searching etc. In this course, advanced methods in the context of dimension reduction, feature extraction and selection, clustering, and classification will be explored. Along with conventional machine learning algorithms, artificial neural networks and computational intelligence methods will also be discussed. The interrelationship between these methods will be addressed, and the mixture-of expert approach will be examined. We will also look at a few case studies on building powerful, intelligent data mining systems. The course is taught with formal and informal lectures and in-class discussion is encouraged. Students will give presentations based on selected research papers of interest. A major assessment component is a project that aims at developing an intelligent data analysis system for real-world problem solving.
A thorough understanding of Linear Algebra and Vector Calculus (e.g., students should be able to easily compute gradients/Hessians of a multivariate function), as well as basic understanding of the Python or MATLAB/OCTAVE programming environments.
20h lecturer, 30h exercise, 15h homework
The final grade will be determined based on regular homeworks, one midterm exam, and a Semester Project:
Homeworks: 20%
Midterm Exam: 30%
Semester Project: 50%
Hands-on design projects are the key component of the course. Team work is required for the projects.
The lectures will follow, in part, Tom Mitchell, Machine Learning, McGraw Hill, 1997. The more advanced material will be based on material the instructor will make available. Some interesting books for the advanced material include:
Major links from Clayton Scott:
Books
Other machine learning courses
Data repositories
Background
Matlab Software
Conferences/Publications
Nearest Neighbors
Density Estimation
Linear methods for classification
Decision Trees
Error estimation
Boosting
Support Vector Machines
Clustering
Dimensionality reduction
Nonlinear regression and Gaussian Processes