Akademik Dersler - İçerik

1. Describe the fundamental concepts and architectures of deep learning. 1.1. Define the basic structures and components of artificial neural networks. 1.2. Interpret the operating principles of MLP, RNN/LSTM, and CNN architectures. 1.3. Develop and implement a simple deep learning model.
2. Identify the fundamental principles of reinforcement learning. 2.1. Define the core concepts and terminology of reinforcement learning. 2.2. Distinguish reinforcement learning from supervised learning approaches.
3. Analyze value-based reinforcement learning methods. 3.1. Describe the Q-learning algorithm and its learning mechanism. 3.2. Interpret the Deep Q-Network (DQN) approach. 3.3. Implement and evaluate a basic DQN agent.
4. Examine policy-based and actor-critic reinforcement learning methods. 4.1. Describe the principles of Policy Gradient methods. 4.2. Analyze the REINFORCE algorithm and its application. 4.3. Interpret the Actor-Critic framework and its components.
5. Compare advanced deep reinforcement learning algorithms. 5.1. Compare PPO, SAC, and DDPG algorithms in terms of structure and application. 5.2. Evaluate the advantages and limitations of advanced reinforcement learning algorithms.
6. Assess exploration strategies and intrinsic motivation mechanisms in reinforcement learning. 6.1. Interpret the exploration–exploitation trade-off in reinforcement learning. 6.2. Describe intrinsic motivation and curiosity-driven learning approaches.
7. Evaluate engineering applications and multi-agent approaches in reinforcement learning. 7.1. Describe engineering applications of reinforcement learning. 7.2. Compare different multi-agent reinforcement learning structures. 7.3. Analyze an engineering problem within the reinforcement learning framework.