NeuralNets Adv Deep Learning

3 Credits · Level: Graduate

Neural Networks and Advanced Deep Learning - Neural Networks and Advanced Deep Learning explores both foundational and cutting-edge deep learning techniques. Students start with a short review of core concepts such as feed-forward networks, activation functions, backpropagation, and optimization using TensorFlow and Keras. The course reviews convolutional and recurrent neural networks, auto-encoders, and methods to prevent overfitting. Advanced topics include Transformers and attention mechanisms, Graph Neural Networks, self-supervised and contrastive learning, neural architecture search, adversarial robustness, energy-based models, and neural ODEs. Additionally, the curriculum delves into advanced optimization, neuro-symbolic AI, and bio-inspired deep learning. Various practical applications are covered to equip students with the skills to address complex deep learning challenges across diverse domains.