Day 1: Introduction to deep learning and neural networks
- Basics of machine learning
- Introduction to neural networks and their functions
- Training neural networks using backpropagation
- Activation functions and their selection
- Multilayer neural networks and their training
- Introduction to TensorFlow and Keras library
Day 2: Convolutional Neural Networks (CNN)
- Introduction to convolutional neural networks (CNN)
- Convolution layers and image filtering
- Layer pooling and reducing image dimensions
- Editing images before training
- Convolutional network architectures (AlexNet, VGG, ResNet)
- Practical exercises for the implementation of convolutional networks
Day 3: Recurrent neural networks (RNN) and other models
- Introduction to recurrent neural networks (RNN)
- Structure and functioning of RNN
- RNN training for text prediction and generation
- LSTM and GRU networks
- Introduction to generative models (GAN)
- Introduction to autoencoders and their applications
Day 4: Optimization and tuning of neural networks
- Optimization of hyperparameters of neural networks
- Methods of learning and parameter optimization
- Regularization of networks (dropout, L1, L2)
- Overfitting and its prevention
- Tuning and debugging networks
Day 5: Applications and advanced topics in deep learning
- Application of deep learning in practice (speech recognition, image recognition, machine translation)
- Transfer learning
- Introduction to natural language processing and NLP
- Advanced topics in deep learning (adversarial training, attention mechanism, capsule networks)
- Discussions and course summary
