DERSLERİN İÇERİKLERİ
(İNGİLİZCE -TEZLİ)
ARI5001 Principles of Artificial Intelligence (3+0) 3 Kredi AKTS:8
This course provides a general introduction to the basic priniples and applications of artificial intelligence. The course is mainly about how to design and implement intelligent agents. The students will have an understanding of the general problem techniques in artificial intelligence, search methods, knowledge representation, uncertainty management, reasoning, planning, perception and action, learning, and artificial intelligence languages. The students will gain knowledge about how to apply these techniques to tackle with new artificial intelligence problems. Also, they will be able to design and implement key components of intelligent agents of moderate complexity.
ARI5003 Machine Learning (3+0) 3 Kredi AKTS:8
This course provides in-depth coverage of machine learning algorithms. The topics that will be included are data pre-processing techniques, model selection and generalization in machine learning, supervised learning problems, Bayesian decision theory, parametric and non-paramteric classification, decision trees, instance-based learning algorithms, density estimation techniques, unsupervised learning problems, dimensionality reduction, and design and analysis of machine learning experiments. Upon the completion of this course, the students will be able to design a machine learning experiment, choose the suitable pre-processing and learning techniques for the given problem, apply them using a programming language and evaluate the results with proper evaluation metrics.
ARI5002 Optimization for Artificial Intelligence (3+0) 3 Kredi AKTS:8
This course will deliver fundamental mathematics and algorithms for optimization approaches to develop and improve Artificial Intelligence solutions. Topics of the course will cover fundamental Optimization Methods, Programming Techniques (mathematical, constraint, graph), Large Scale Optimization Methods, Distributed Optimization, Optimization Strageties for Deep Learning and Parallel Optimization Methods, advantages and disadvantages, applications of optimization methods in real world problems.
ARI5887 Seminar (0+0) 0 Kredi AKTS:4
Seminars are given by instructors, invited speakers and students enrolled in the course. Student presentations may also include thesis studies. In the course students will be informed about the things to consider in giving a successful presentation.
EEE5101 Research Methods and Ethics (1+0)1 Kredi AKTS:2
This course aims to provide graduate students with endowments to academically communicate their researches in an written and oral language. It provides them with comprehensive overview of various research skills that balance the practicalities of conducting research and the theory and debates that keep qualitative inquiry vibrant.It covers the entire research process including: formulating research questions; sampling (probability and nonprobability); measurement (surveys, scaling, qualitative, unobtrusive); research design (experimental and quasi-experimental); data analysis; and, writing the research paper. It also addresses the major theoretical and philosophical underpinnings of research including: the idea of validity in research, reliability of measures, scientific research ethics, and publication ethics.
ARI5888-I Master Thesis (0+0) 0 Kredi AKTS:30
This is a thesis study of the student conducted by an academic advisor in the subject of Artificial Intelligence.
ARI5888-II Master Thesis (0+0) 0 Kredi AKTS:30
This is a thesis study of the student conducted by an academic advisor in the subject of Artificial Intelligence.
YAPAY ZEKA YÜKSEK LİSANS PROGRAMI
ZORUNLU DERSLERİN İÇERİKLERİ
(İNGİLİZCE -TEZSİZ)
ARI5001 Principles of Artificial Intelligence (3+0) 3 Kredi AKTS:8
This course provides a general introduction to the basic priniples and applications of artificial intelligence. The course is mainly about how to design and implement intelligent agents. The students will have an understanding of the general problem techniques in artificial intelligence, search methods, knowledge representation, uncertainty management, reasoning, planning, perception and action, learning, and artificial intelligence languages. The students will gain knowledge about how to apply these techniques to tackle with new artificial intelligence problems. Also, they will be able to design and implement key components of intelligent agents of moderate complexity.
ARI5003 Machine Learning (3+0) 3 Kredi AKTS:8
This course provides in-depth coverage of machine learning algorithms. The topics that will be included are data pre-processing techniques, model selection and generalization in machine learning, supervised learning problems, Bayesian decision theory, parametric and non-paramteric classification, decision trees, instance-based learning algorithms, density estimation techniques, unsupervised learning problems, dimensionality reduction, and design and analysis of machine learning experiments. Upon the completion of this course, the students will be able to design a machine learning experiment, choose the suitable pre-processing and learning techniques for the given problem, apply them using a programming language and evaluate the results with proper evaluation metrics.
ARI5002 Optimization for AI (3+0) 3 Kredi AKTS:8
This course will deliver fundamental mathematics and algorithms for optimization approaches to develop and improve Artificial Intelligence solutions. Topics of the course will cover fundamental Optimization Methods, Programming Techniques (mathematical, constraint, graph), Large Scale Optimization Methods, Distributed Optimization, Optimization Strageties for Deep Learning and Parallel Optimization Methods, advantages and disadvantages, applications of optimization methods in real world problems.
ARI5004 Deep Learning (3+0) 3 Kredi AKTS:8
The course provides in-depth coverage of the Deep Leanring concepts and most recent neural network architectures and algorithms. The topics inlude simple perceptron, backpropagation algorithm, multi-layer perceptron, radial basis functions, mixture of experts, recurrent neural networks, long short-term memory networks, convolutional neural networks, reinforcement learning, transfer learning and fine-tuning techniques, autoencoder networks, generative adversarial networks, and regularization techniques. Students will also be familiar with the most recent applications of deep learning algorithms. Upon the completion of this course, students will be able to to use neural networks in order to solve real world artificial intelligence problems.
EEE5101 Research Methods and Ethics (1+0)1 Kredi AKTS:2
This course aims to provide graduate students with endowments to academically communicate their researches in an written and oral language. It provides them with comprehensive overview of various research skills that balance the practicalities of conducting research and the theory and debates that keep qualitative inquiry vibrant.It covers the entire research process including: formulating research questions; sampling (probability and nonprobability); measurement (surveys, scaling, qualitative, unobtrusive); research design (experimental and quasi-experimental); data analysis; and, writing the research paper. It also addresses the major theoretical and philosophical underpinnings of research including: the idea of validity in research, reliability of measures, scientific research ethics, and publication ethics.
ARI5999 Project (0+0) 0 Kredi AKTS:12
This course is a project study of the student conducted by an academic advisor on the subject of Artificial Intelligence.
YAPAY ZEKA YÜKSEK LİSANS PROGRAMI
SEÇMELİ DERSLERİN İÇERİKLERİ
(İNGİLİZCE-TEZLİ)
ARI5004 Deep Learning (3+0) 3 Kredi AKTS:8
The course provides in-depth coverage of the Deep Leanring concepts and most recent neural network architectures and algorithms. The topics inlude simple perceptron, backpropagation algorithm, multi-layer perceptron, radial basis functions, mixture of experts, recurrent neural networks, long short-term memory networks, convolutional neural networks, reinforcement learning, transfer learning and fine-tuning techniques, autoencoder networks, generative adversarial networks, and regularization techniques. Students will also be familiar with the most recent applications of deep learning algorithms. Upon the completion of this course, students will be able to to use neural networks in order to solve real world artificial intelligence problems.
ARI5510 Hardware Technologies for AI (3+0) 3 Kredi AKTS:12
This course provides two-way approach at the intersection of Hardware systems and artificial intelligence solutions. First, a general introduction to hardware architectures and design techniques used to design fast and efficient training and inference algorithms in artificial intelligence will be covered. Then, then classical ML algorithms as well as DNN models from the perspective of the impact of parameters such as batch size, precision, sparsity and compression on the accuracy of these models will be studied. General hardware implementation techniques will be introduced for using parallelism, locality, and low precision to implement the core computational kernels to build the relationships between model parameters and hardware implementation techniques. Overall, algorithmic approaches that could be mapped onto hardware for better speed and energy consumption, and hardware design
key principles to support Deep Learning architectures will be the main streams of this course.
ARI5012 Cloud Computing (3+0) 3 Kredi AKTS:12
This course includes the understanding and usage of cloud computing and its related technologies (Virtual Servers, SAAS, IAAS, cloud based networks, cloud based databases), Defining Cloud Computing solutions, establishing connections with large data infrastructures, combining Cloud Computing solutions with best practices and large data.
ARI5501 Natural Language Processing (3+0) 3 Kredi AKTS:12
The course introduces the students into the principles and methods of natural language processing, with a focus on current methods and technologies. The course additionally aims at an understanding of the underlying computational properties of natural language and of current research directions. We study core tasks in natural language processing, including language modeling, syntactic analysis, semantic interpretation, coreference resolution, discourse analysis, dialogue modeling and machine translation. We discuss the underlying linguistic phenomena, i.e., the linguistic features, and how they can be modeled and automatically learned from data using deep learning techniques. We illustrate the methods and technologies with current applications in real world settings.
CMP5550 Computer Vision (3+0) 3 Kredi AKTS:8
This class introduces the fundamental techniques in computer vision. Initially basic concepts of image formation, representation and camera projection geometries will be given. Later some classical image processing techniques will be introduced such as edge detection, segmentation, thresholding etc. Image matching, optical flow, local image features will be described in the context of multiple image processing. Basic image recognition techniques are also to be introduced. 3D inference will be another focus where stereo imaging, 3D reconstruction and various shape from X techniques are to be discussed.
CMP6160 Advanced Topics in Artificial Intelligence (3+0) 3 Kredi AKTS:9
This is a research oriented course. Topics vary from one offering to the next. Selected state of the art research papers from various fields of artificial intelligence (such as searching, knowledge representation, learning, probabilistic reasoning, and natural language processing) will be read. Students are required to write paper reviews and do a final project.
CMP5208 Paralel Computing with GPUs (3+0) 3 Kredi AKTS:12
This course provides an introduction to parallel computing with emphasis on programming massively parallel processor such as graphics processor units (GPUs). The students will make extensive use of parallel programming schemes such as Compute Unified Device Architecture (CUDA). The topics covered are instruction and data level parallelism, CUDA programming, control flow and synchronization, shared memory programming, performance optimization. The students will also gain familiarity with the GPU programming tools.
CMP5203 High-Performance Computer Architecture (3+0)3 Kredi AKTS:12
This course overviews graduate-level computer architecture concepts. The course builds on fundamental processor and memory design issues by in depth analysis of high-performance computer systems. The course aims at exploring advanced level parallelism concepts such as instruction-level parallelism and data-level parallelism. Further, the course provides insights in recent trends in high-performance computing such as GPU architectures and warehouse-scale computers. Also, the students will gain familiarity with new trends in computing such as cloud, big data, and internet of things.
CMP5132 Social Network Analysis (3+0) 3 Kredi AKTS:12
The main objective of this course is for students to gain hands-on experiences on Social Network Analysis. Course covers techniques used to collect, analyze, and understand the data from Internet and Online social networks. At the end of the class, students should be able to understand the whole process of collecting information from the web, and carrying out system design for search and mining the social network websites (facebook, twitter, etc.). We will use online web documents (such as Twitter data) as the test set and practice social network mining techniques.
CMP5101 Data Mining I (3+0) 3 Kredi AKTS:8
This course provides an introduction to data mining concepts. Basic concepts in data mining: frequent item set detection, association rules, clustering and classification are covered in depth. The students will learn how to approach data mining as a process, by demonstrating competency in the use of CRISP-DM, the Cross-Industry Standard Process or Data Mining, including the business understanding phase, the data understanding phase, the exploratory data analysis phase, the modeling phase, the evaluation phase, and the deployment phase. Also, they will understand and apply the most current data mining techniques and applications, such as text mining, mining genomics data, and other current issues.
CMP5102 Data Mining II (3+0) 3 Kredi AKTS:12
This course introduces some advanced and popular data mining topics with practical implementations. Applications shall be made on the R open-source software program. Basic information will be taught to use R programming language. The content of this course includes Data import and export, Data exploration, Decision trees and random forest, Network estimation, Outlier detection, Time series analysis, Association rules, Text mining, Social network analysis, web mining, Case study I: Analysis and forecasting of house price indices, Case study II: Predictive modelling of Big Data with limited memory.
CMP5126 Image and Video Processing (3+0) 3 Kredi AKTS:12
The objective of this course is to introduce the basics of image and video processing methodologies. The main topics of this course are; Image and video processing concepts and applications; Introduction to multidimensional signal processing: sampling, filtering, interpolation and truncation; Human visual perception; Scanning and displaying images and video; Image enhancement, restoration and segmentation; Digital image and video compression; It is in the form of image analysis methods.
Cmp5204 Embedded Systems (3+0) 3 Kredi AKTS:12
This course is a hands-on course that requires writing software as well as board-level work. It sits at the intersection of fields such as microprocessors, digital design, operating systems, software design, and industrial automation. The students are exposed to topics such as meeting real-time constraints in embedded systems, generating delays and interrupts, using the serial interface, etc. They get theoretical as well as hands-on experience on embedded system design by using embedded software development environments and hardware emulators, as well as by working on actual hardware where they physically connect multiple building blocks.
BDA5001 Introduction to Big Data (3+0) 3 Kredi AKTS:8
This course is designed to improve the analytical skills of students for business decision making and will provide insights into the basics of using the "Big Data" to quantify the operational impact of management decisions. This course includes modeling techniques, advanced data management, data visualization, optimization, risk analysis and simulation modeling.
BDA5002 Marketing Analytics (3+0) 3 Kredi AKTS:8
This course focuses on marketing analytics methods and applications that are used to develop marketing strategies, and create a link between marketing, customer behavior and business outcome. In this course, concepts, methods and applications of decision modeling will be studied. An analytical approach will be presented to topics such as market segmentation, targeting, positioning, pricing and promotional planning. The class instruction and discussion will cover some theory and methods and various statistical methods and software will be used to address practical business problems and/or case studies. At the end of this course, students will become familiar with key existing and emerging marketing issues, and research/analytics approaches to address them from a practical perspective.
SEN5550 Business Intelligence (3+0) 3 Kredi AKTS:8
The content of this course is composed of introduction to business intelligence, database management systems, data warehouse models and architectures, data mining, preprocessing, driven methodology, guided algorithms and non-guided algorithms. Participants will learn the usage of business intelligence, the contribution of data mining methods in business intelligence. They will be introduced to open-source and commercial business intelligence solutions, and develop business intelligence applications.
YAPAY ZEKA YÜKSEK LİSANS PROGRAMI
SEÇMELİ DERSLERİN İÇERİKLERİ
(İNGİLİZCE-TEZSİZ)
ARI5510 Hardware Technologies for AI (3+0) 3 Kredi AKTS:8
This course provides two-way approach at the intersection of Hardware systems and artificial intelligence solutions. First, a general introduction to hardware architectures and design techniques used to design fast and efficient training and inference algorithms in artificial intelligence will be covered. Then, then classical ML algorithms as well as DNN models from the perspective of the impact of parameters such as batch size, precision, sparsity and compression on the accuracy of these models will be studied. General hardware implementation techniques will be introduced for using parallelism, locality, and low precision to implement the core computational kernels to build the relationships between model parameters and hardware implementation techniques. Overall, algorithmic approaches that could be mapped onto hardware for better speed and energy consumption, and hardware design
key principles to support Deep Learning architectures will be the main streams of this course.
ARI5012 Cloud Computing (3+0) 3 Kredi AKTS:12
This course includes the understanding and usage of cloud computing and its related technologies (Virtual Servers, SAAS, IAAS, cloud based networks, cloud based databases), Defining Cloud Computing solutions, establishing connections with large data infrastructures, combining Cloud Computing solutions with best practices and large data.
ARI5501 Natural Language Processing (3+0) 3 Kredi AKTS:12
The course introduces the students into the principles and methods of natural language processing, with a focus on current methods and technologies. The course additionally aims at an understanding of the underlying computational properties of natural language and of current research directions. We study core tasks in natural language processing, including language modeling, syntactic analysis, semantic interpretation, coreference resolution, discourse analysis, dialogue modeling and machine translation. We discuss the underlying linguistic phenomena, i.e., the linguistic features, and how they can be modeled and automatically learned from data using deep learning techniques. We illustrate the methods and technologies with current applications in real world settings.
CMP5550 Computer Vision (3+0) 3 Kredi AKTS:8
This class introduces the fundamental techniques in computer vision. Initially basic concepts of image formation, representation and camera projection geometries will be given. Later some classical image processing techniques will be introduced such as edge detection, segmentation, thresholding etc. Image matching, optical flow, local image features will be described in the context of multiple image processing. Basic image recognition techniques are also to be introduced. 3D inference will be another focus where stereo imaging, 3D reconstruction and various shape from X techniques are to be discussed.
CMP6160 Advanced Topics in Artificial Intelligence (3+0) 3 Kredi AKTS:9
This is a research oriented course. Topics vary from one offering to the next. Selected state of the art research papers from various fields of artificial intelligence (such as searching, knowledge representation, learning, probabilistic reasoning, and natural language processing) will be read. Students are required to write paper reviews and do a final project.
CMP5208 Paralel Computing with GPUs (3+0) 3 Kredi AKTS:12
This course provides an introduction to parallel computing with emphasis on programming massively parallel processor such as graphics processor units (GPUs). The students will make extensive use of parallel programming schemes such as Compute Unified Device Architecture (CUDA). The topics covered are instruction and data level parallelism, CUDA programming, control flow and synchronization, shared memory programming, performance optimization. The students will also gain familiarity with the GPU programming tools.
CMP5203 High-Performance Computer Architecture (3+0)3 Kredi AKTS:12
This course overviews graduate-level computer architecture concepts. The course builds on fundamental processor and memory design issues by in depth analysis of high-performance computer systems. The course aims at exploring advanced level parallelism concepts such as instruction-level parallelism and data-level parallelism. Further, the course provides insights in recent trends in high-performance computing such as GPU architectures and warehouse-scale computers. Also, the students will gain familiarity with new trends in computing such as cloud, big data, and internet of things.
CMP5132 Social Network Analysis (3+0) 3 Kredi AKTS:12
The main objective of this course is for students to gain hands-on experiences on Social Network Analysis. Course covers techniques used to collect, analyze, and understand the data from Internet and Online social networks. At the end of the class, students should be able to understand the whole process of collecting information from the web, and carrying out system design for search and mining the social network websites (facebook, twitter, etc.). We will use online web documents (such as Twitter data) as the test set and practice social network mining techniques.
CMP5101 Data Mining I (3+0) 3 Kredi AKTS:8
This course provides an introduction to data mining concepts. Basic concepts in data mining: frequent item set detection, association rules, clustering and classification are covered in depth. The students will learn how to approach data mining as a process, by demonstrating competency in the use of CRISP-DM, the Cross-Industry Standard Process or Data Mining, including the business understanding phase, the data understanding phase, the exploratory data analysis phase, the modeling phase, the evaluation phase, and the deployment phase. Also, they will understand and apply the most current data mining techniques and applications, such as text mining, mining genomics data, and other current issues.
CMP5102 Data Mining II (3+0) 3 Kredi AKTS:12
This course introduces some advanced and popular data mining topics with practical implementations. Applications shall be made on the R open-source software program. Basic information will be taught to use R programming language. The content of this course includes Data import and export, Data exploration, Decision trees and random forest, Network estimation, Outlier detection, Time series analysis, Association rules, Text mining, Social network analysis, web mining, Case study I: Analysis and forecasting of house price indices, Case study II: Predictive modelling of Big Data with limited memory.
CMP5126 Image and Video Processing (3+0) 3 Kredi AKTS:12
The objective of this course is to introduce the basics of image and video processing methodologies. The main topics of this course are; Image and video processing concepts and applications; Introduction to multidimensional signal processing: sampling, filtering, interpolation and truncation; Human visual perception; Scanning and displaying images and video; Image enhancement, restoration and segmentation; Digital image and video compression; It is in the form of image analysis methods.
Cmp5204 Embedded Systems (3+0) 3 Kredi AKTS:12
This course is a hands-on course that requires writing software as well as board-level work. It sits at the intersection of fields such as microprocessors, digital design, operating systems, software design, and industrial automation. The students are exposed to topics such as meeting real-time constraints in embedded systems, generating delays and interrupts, using the serial interface, etc. They get theoretical as well as hands-on experience on embedded system design by using embedded software development environments and hardware emulators, as well as by working on actual hardware where they physically connect multiple building blocks.
BDA5001 Introduction to Big Data (3+0) 3 Kredi AKTS:8
This course is designed to improve the analytical skills of students for business decision making and will provide insights into the basics of using the "Big Data" to quantify the operational impact of management decisions. This course includes modeling techniques, advanced data management, data visualization, optimization, risk analysis and simulation modeling.
BDA5002 Marketing Analytics (3+0) 3 Kredi AKTS:8
This course focuses on marketing analytics methods and applications that are used to develop marketing strategies, and create a link between marketing, customer behavior and business outcome. In this course, concepts, methods and applications of decision modeling will be studied. An analytical approach will be presented to topics such as market segmentation, targeting, positioning, pricing and promotional planning. The class instruction and discussion will cover some theory and methods and various statistical methods and software will be used to address practical business problems and/or case studies. At the end of this course, students will become familiar with key existing and emerging marketing issues, and research/analytics approaches to address them from a practical perspective.
SEN5550 Business Intelligence (3+0) 3 Kredi AKTS:8
The content of this course is composed of introduction to business intelligence, database management systems, data warehouse models and architectures, data mining, preprocessing, driven methodology, guided algorithms and non-guided algorithms. Participants will learn the usage of business intelligence, the contribution of data mining methods in business intelligence. They will be introduced to open-source and commercial business intelligence solutions, and develop business intelligence applications.
EN
