Akademik Dersler - Tanıtım


Code - Course Title	BİM4501 - Deep Reinforcement Learning
Course Type	Area Elective Courses
Language of Instruction	İngilizce
Laboratory + Practice	3+0
ECTS	4.5
Course Instructor(s)	ÖĞRETİM GÖREVLİSİ Enver BİLDİK
Mode of Delivery	Face to face
Prerequisites	There are no prerequisites for this course
Courses Recomended	Students who take this course are recommended to have previously completed BİM447 – Introduction to Deep Learning.
Required or Recommended Resources	Richard S. Sutton, Andrew G. Barto – Reinforcement Learning: An Introduction (2nd Edition)Maxim Lapan – Deep Reinforcement Learning Hands-On (3rd Edition)Phil Winder – Reinforcement Learning: Industrial Applications of Intelligent Agents
Recommended Reading List	https://huggingface.co/learn/deep-rl-course/unit0/introductionhttps://gymnasium.farama.org/https://spinningup.openai.com/en/latest/
Assessment methods and criteria	There will be one midterm exam, one project, and one final exam. The exams will be in a written (essay-style) format.
Work Placement	There is no internship or practical training component for this course.
Sustainability Development Goals

Weeks	Topics
Week - 1	Introduction to Deep Neural Networks
Week - 2	Recurrent Neural Networks (RNN)
Week - 3	Convolutional Neural Networks (CNN)
Week - 4	Introduction to Reinforcement Learning (RL)
Week - 5	Deep Q-Networks (DQN),
Week - 6	Policy Gradient Methods
Week - 7	Midterm Exam
Week - 8	Actor-Critic Methods
Week - 9	Advanced Deep RL Algorithms-I
Week - 10	Advanced Deep RL Algorithms-II
Week - 11	Exploration and Intrinsic Motivation
Week - 12	Deep RL in Robotics and Control
Week - 13	Multi-Agent Reinforcement Learning (MARL)
Week - 14	Final Exam

Assessment Method and Passing Requirements
	Quamtity	Percentage (%)
Toplam (%)		0