Code CS 412
Term 201801
Title Machine Learning
Faculty Faculty of Engineering and Natural Sciences
Subject Computer Sci.& Eng.(CS)
SU Credit 3
ECTS Credit 10.00 / 6.00 ECTS (for students admitted in the 2013-14 Academic Year or following years)
Instructor(s) Ayse Berrin Yan?ko?lu berrin@sabanciuniv.edu,
Detailed Syllabus
Language of Instruction English
Prerequisites
(only for SU students)
MATH203 MATH201
Mode of Delivery Formal lecture,Recitation
Planned Learning Activities Interactive,Learner centered,Project based learning
Content

This is an introductory machine learning course that will aim a solid understanding of the fundamental issues in machine learning (overfitting, bias/variance), together with several state-of-art approaches such as decision trees, linear regression, k-nearest neighbor, Bayesian classifiers, neural networks, logistic regression, and classifier combination.

Objective

To teach fundamentals of Machine Learning for students of all backgrounds, so that they will know its capabilities, limitations and be able to design all aspects of a learning system.

Learning Objectives:
1. Understand the basic concepts, issues, assumptions and limitations in machine learning (e.g. base accuracy, overfitting, bias/variance, curse of dimensionality...).
2. Have a working knowledge of the basic mathematics (probability, expectation, entropy, basic linear algebra, ...) and algorithms behind common machine learning techniques; together with their suitability in given situations.
3. Given a machine learning problem, be able to implement and evaluate one of the standard machine learning algorithms (e.g. decision trees, neural networks, SVMs) using a tool such as Weka/Matlab or a programming language Python/R.

Learning Outcome

Have a solid understanding of the basic concepts, issues, assumptions and limitations in machine learning and how they apply to various machine learning techniques.
Have a working knowledge of the basic mathematics (probability, expectation, entropy, basic linear algebra, ...) and algorithms behind common machine learning techniques; together with their suitability in given situations.
Given a machine learning problem, select, implement and evaluate one of the appropriate machine learning algorithms (e.g. decision trees, neural networks, SVMs) using a programming environment such as Weka or Matlab.

Programme Outcomes

 Common Outcomes For All Programs 1 Understand the world, their country, their society, as well as themselves and have awareness of ethical problems, social rights, values and responsibility to the self and to others. 1 2 Understand different disciplines from natural and social sciences to mathematics and art, and develop interdisciplinary approaches in thinking and practice. 1 3 Think critically, follow innovations and developments in science and technology, demonstrate personal and organizational entrepreneurship and engage in life-long learning in various subjects. 3 4 Communicate effectively in Turkish and English by oral, written, graphical and technological means. 3 5 Take individual and team responsibility, function effectively and respectively as an individual and a member or a leader of a team; and have the skills to work effectively in multi-disciplinary teams. 3 Common Outcomes ForFaculty of Eng. & Natural Sci. 1 Possess sufficient knowledge of mathematics, science and program-specific engineering topics; use theoretical and applied knowledge of these areas in complex engineering problems. 5 2 Identify, define, formulate and solve complex engineering problems; choose and apply suitable analysis and modeling methods for this purpose. 5 3 Develop, choose and use modern techniques and tools that are needed for analysis and solution of complex problems faced in engineering applications; possess knowledge of standards used in engineering applications; use information technologies effectively. 5 4 Ability to design a complex system, process, instrument or a product under realistic constraints and conditions, with the goal of fulfilling specified needs; apply modern design techniques for this purpose. 4 5 Design and conduct experiments, collect data, analyze and interpret the results to investigate complex engineering problems or program-specific research areas. 5 6 Knowledge of business practices such as project management, risk management and change management; awareness on innovation; knowledge of sustainable development. 1 7 Knowledge of impact of engineering solutions in a global, economic, environmental, health and societal context; knowledge of contemporary issues; awareness on legal outcomes of engineering solutions; understanding of professional and ethical responsibility. 2 Common Outcomes ForSchool of Management 1 Demonstrate an understanding of economics, and main functional areas of management. 1 2 Assess the impact of the economic, social, and political environment from a global, national and regional level. 1 Computer Science and Engineering Program Outcomes Core Electives 1 Design, implement, test, and evaluate a computer system, component, or algorithm to meet desired needs and to solve a computational problem. 5 2 Demonstrate knowledge of discrete mathematics and data structures. 1 3 Demonstrate knowledge of probability and statistics, including applications appropriate to computer science and engineering. 5 Mechatronics Engineering Program Outcomes Core Electives 1 Familiarity with concepts in statistics and optimization, knowledge in basic differential and integral calculus, linear algebra, differential equations, complex variables, multi-variable calculus, as well as physics and computer science, and ability to use this knowledge in modeling, design and analysis of complex dynamical systems containing hardware and software components. 5 2 Ability to work in design, implementation and integration of engineering applications, such as electronic, mechanical, electromechanical, control and computer systems that contain software and hardware components, including sensors, actuators and controllers. 5 Electronics Engineering Program Outcomes Area Electives 1 Use mathematics (including derivative and integral calculations, probability and statistics), basic sciences, computer and programming, and electronics engineering knowledge to design and analyze complex electronic circuits, instruments, software and electronics systems with hardware/software. 5 2 Analyze and design communication networks and systems, signal processing algorithms or software using advanced knowledge on differential equations, linear algebra, complex variables and discrete mathematics. 5 Molecular Biology, Genetics and Bioengineering Program Outcomes Area Electives 1 Comprehend key concepts in biology and physiology, with emphasis on molecular genetics, biochemistry and molecular and cell biology as well as advanced mathematics and statistics. 1 2 Develop conceptual background for interfacing of biology with engineering for a professional awareness of contemporary biological research questions and the experimental and theoretical methods used to address them. 1 Industrial Engineering Program Outcomes Area Electives 1 Formulate and analyze problems in complex manufacturing and service systems by comprehending and applying the basic tools of industrial engineering such as modeling and optimization, stochastics, statistics. 5 2 Design and develop appropriate analytical solution strategies for problems in integrated production and service systems involving human capital, materials, information, equipment, and energy. 2 3 Implement solution strategies on a computer platform for decision-support purposes by employing effective computational and experimental tools. 2 Materials Science and Nano Engineering Program Outcomes Area Electives 1 Applying fundamental and advanced knowledge of natural sciences as well as engineering principles to develop and design new materials and establish the relation between internal structure and physical properties using experimental, computational and theoretical tools. 2 2 Merging the existing knowledge on physical properties, design limits and fabrication methods in materials selection for a particular application or to resolve material performance related problems. 1 3 Predicting and understanding the behavior of a material under use in a specific environment knowing the internal structure or vice versa. 1
 Assessment Methods and Criteria Percentage (%) Final 35 Midterm 30 Group Project 15 Homework 20
 Recommended or Required Reading Textbook Intro. to Machine Learning - Ethem Alpayd?n Optional Readings Advanced or alternative explanations will be provided as supplement to lecture slides and course textbook. Course Web http://people.sabanciuniv.edu/berrin/cs512/