Akademik Dersler - Tanıtım


Code - Course Title	ENM529 - Optimization for Machine Learning
Course Type	Elective Courses
Language of Instruction	Türkçe
Laboratory + Practice	3+0
ECTS	7.5
Course Instructor(s)	DOÇENT DOKTOR GÜLÇİN DİNÇ YALÇIN
Mode of Delivery	in person
Prerequisites	-
Courses Recomended	-
Required or Recommended Resources	• Sra, S., Nowozin, S., & Wright, S. J. (Eds.). (2012). Optimization for machine learning. Mit Press.• Schmidt, M., Le Roux, N., & Bach, F. (2017). Minimizing finite sums with the stochastic average gradient. Mathematical Programming, 162, 83-112.• Newton, D., Yousefian, F., & Pasupathy, R. (2018). Stochastic gradient descent: Recent trends. In INFORMS TutORials in Operations Research. Published online: 19 Oct 2018; 193-220.• Bottou, L., Curtis, F. E., & Nocedal, J. (2018). Optimization methods for large-scale machine learning. SIAM review, 60(2), 223-311.• Netrapalli, P. (2019). Stochastic gradient descent and its variants in machine learning. Journal of the Indian Institute of Science, 99(2), 201-213.
Recommended Reading List	-
Assessment methods and criteria	midterm exam, project, final exam
Work Placement	-
Sustainability Development Goals

Weeks	Topics
Week - 1	Optimization, machine learning and convexity
Week - 2	Convex optimization for machine learning
Week - 3	General formulation of machine learning problems and machine learning problem: Support Vector Machines
Week - 4	Machine learning problem: Support Vector Machines
Week - 5	Machine learning problem: Logistic Regression
Week - 6	Gradient decent method
Week - 7	Line search in gradient descent method
Week - 8	Stochastic gradient descent method
Week - 9	Noise reduction methods
Week - 10	Adaptive methods
Week - 11	Convergence analysis of first-order methods
Week - 12	Second-order methods
Week - 13	Stochastic BFGS and L-BFGS methods
Week - 14	Convergence analysis of second-order methods

Assessment Method and Passing Requirements
	Quamtity	Percentage (%)
1.Midterm Exam	1	30
Homework	1	30
Final Exam	1	40
Toplam (%)		100