Akademik Dersler - İçerik

1.Introduction to Artificial Neural Networks 1. Students will be able to define the basic building blocks of artificial neural networks (neurons, weights, biases). 2. Students will understand the structure of perceptrons and feed-forward networks. 3. Students will be able to build and run a simple neural network model. 2. Training and Evaluation of Neural Networks 1. Students will be able to explain types of activation functions and their effects. 2. Students will evaluate network performance by understanding the train-test split. 3. Students will train networks using the backpropagation algorithm. 3. Foundations of Deep Learning 1. Students will analyze the layered structure of deep neural architectures. 2. Students will identify hyperparameters and discuss their impact on learning. 3. Students will apply regularization techniques (e.g., dropout, L2) to address overfitting. 4. Optimization and Performance in Deep Learning 1. Students will compare the advantages and disadvantages of various optimization algorithms. 2. Students will effectively tune learning rate and other hyperparameters. 3. Students will select and use appropriate evaluation metrics to measure model performance. 5. Convolutional Neural Networks (CNN) 1. Students will explain convolution operation, filters, and feature maps. 2. Students will describe popular CNN architectures (LeNet, AlexNet, VGG) with examples. 3. Students will demonstrate how CNNs are applied to image processing tasks. 6. Recurrent Neural Networks (RNN) and Time Series 1. Students will explain RNN structure for handling time series data. 2. Students will differentiate advanced RNN architectures like LSTM and GRU. 3. Students will understand training of time-dependent models using BPTT. 7. Fundamentals of Reinforcement Learning 1. Students will explain the agent-environment interaction loop and Markov Decision Processes (MDPs). 2. Students will evaluate the exploration vs. exploitation trade-off with examples. 3. Students will relate key RL concepts (policy, reward, value function) with practical applications.