top of page
본 사이트는
.com
사이트 제작 도구로 제작되었습니다. 지금 나만의 사이트를 만들어보세요.시작하기
Advanced Stochastic Modeling (고등확률모델링)
 
Textbook : 대기행렬이론 (이호우, 시그마프레스)
References : OR/MS 및 확률모형, 이호우 (Sigma Press), Stochastic Processes, S.M. Ross
Course description : This course deals with probability models which is one of the main subjects of operations research. The lecture begins with review of the basic probability theories which include random variables, distribution functions, expectations. Based on the elementary probability models, various random processes will be introduced, among which are Poisson process, renewal process, discrete time and continuous Markov chain, and birth-and-death process. Lastly, queueing theory, one of the most important application area of the probability theory is introduced. Mainly, M/M/․ type and M/G/1 and related models are dealt with in this course. Some models of networks of queues will be studied and their solution procedures are also given, if time permits.
 
Basic Probability Theory
Exponential Distribution and Poisson Process
Renewal Process and Applications
Markov Process and Applications
Queueing Theory and Applications

 

Prerequisite: Elementary calculus. At least one year's probability and statistics course. 
Queueing Theory and Applications (대기이론과 응용)
 
Textbook : 대기행렬이론 (이호우, 시그마프레스), Queueing Networks and Markov Chains (G. Bolch et al.)
Prerequisite : Stochastic processes E
Valuation : H/W Problems 40%, Exams 30%, Presentation 30%
 
확률과정론 Review
단계법 
M/G/1
M/G/1 휴가모형
우선순위 및 기타 서비스 규칙
대기 네트워크
QN Softwares
Data Mining (데이터마이닝)
 
Course description : 데이터마이닝(Data Mining)은 방대하고 복잡한 데이터 내부에 존재하는 유용하고 의미 있는 정보를 이끌어 내는 방법을 연구하는 학문이다. 주로 숫자형태의 일정한 데이터구조로 정형화된 데이터를 분석해 왔으나 최근 들어 비정형 데이터 중에서 웹마이닝과 텍스트마이닝의 중요성이 더욱 부각되고 있다. 본 강의에서는 이러한 데이터마이닝의 기본적인 기법들을 배우고 현실의 여러 문제들을 풀어보는 실습을 통해 기법들의 활용능력을 습득한다.

 

Textbook : Introduction to Data Mining, Ping-Ning Tan, Michael Steinback, Vipin Kumar, Pearson
 

Topics:

Exploring Data

Classification: Basic Concepts, Decision Trees, and Model Evaluation

Classification: Alternative Techniques

Association Analysis: Basic Concepts and Algorithms

Association Analysis: Advanced Concepts

Cluster Analysis: Basic Concepts and Algorithms

Cluster Analysis: Additional Issues and Algorithms

Special Topics in Applied Probabilities (응용확률특론)
 
Course description : 확률이론과 관련된 보다 고급의 이론들 및 응용 가운데서 선택적으로 강의를 한다. 주된 강의내용은 마코프 의사결정론(Markov Decision Processes), 최적정지시간(Optimal stopping times), 재생이론, 마코프 재생과정 등을 다룬다. 주된 응용은 manufacturing 및 정보통신시스템 분야를 다룬다. 또한 강의와 더불어 교재내용을 수강생이 요약 발표하는 세미나식으로 진행하며, 관련 논문들을 찾아 숙지하고 발표하는 형태로 진행한다. (특론 과목이므로, 매번 개설할 때마다 다른 topic을 가지고 강의할 수도 있음) 확률과정론에서는 시스템의 모델링을 목표로 관련 이론들을 알아보았으나 여기서는 모형에 비용구조를 포함시켜 확률과정상의 최적 의사결정을 내리는 기법들을 소개한다.

 

Textbook : Introduction to Operations Research, Hillier and Lieberman Applied Probability and Stochastic Processes, Feldman Introduction to Stochastic Processes, Cinlar Introduction to Stochastic Dynamic Programming, Ross
 

Topics:

Stochastic Inventory:Single period/No setup cost Single period/Setup cost Multi period/No setup cost Multi period/Setup cost
Markov Decision Processes (MDP): Definitions Cost Criterion (Total discounted, Long-run Average) Algorithms LP Formulation Finite-stage model Stock-option model 등
Optimal Stopping TImes: special cases of MDP 
Advanced Models: Markov renewal processes, M/G/1, Game theory
Industrial Artificial Intelligence (산업인공지능)
 
Textbook : 기계학습 (오일석, 한빛아카데미)
References : 
Course description :

Industrial AI is the technique that creates new values by applying AI to various industrial sectors such as manufacturing industry and related forward and backward service industries. Through this lecture, students can understand and study the concept of AI, deep learning with various regulation methods, CNN, RNN, reinforced learning, and unsupervised learning and so forth. Additionally, students perform computer programming jobs that utilize AI techniques to learn how to apply the industrial AI to various industries.

 

- To understand the definition, concept, and the role as a tool to realize the artificial intelligence of the machine learning

- To explain the importance of the model selection and method, overfitting and underfitting

- To learn the data expansion and weight decay which are very important regularization techniques in todays machine kearning

- To introduce supervised learning, unsupervised learning, reinforced learning, semi-supervised learning

 

Specific contents are:

Multi-layer perceptron

Basic deep learning, CNN, deep learning optimization

Unsupervised learning, semi-supervised learning, transfer learning

RNN

Reinforced learning and Q-learning

Kernel method, ensemble method

Application to the manufacturing process

 

Each technique is supplemented by programming with e.g., Python, to be equipped with practical application ability.

 

 

Prerequisite: Elementary calculus. At least one year's probability and statistics course. 
bottom of page