- Understand the design principles of neural networks;
- Understand the concept of activation function;
- Understand the backpropagation algorithm for training a neural network;
- Being able to build a neural network to solve classification tasks;
- Being able to use Keras or similar libraries to build a Neural Network;
- Understand the convolution operator and the idea behind convolutional neural network;
- Understand the main principles of recurrent neural network;
- Understand LSTM and how they can be applied to counteract vanishing gradient problem.
- Being able to apply one of the deep model presented to solve classification or regression tasks.
Weekly - Available soon
Total - Available soon
Portuguese. If there are Erasmus students, classes will be taught in English
Deep Learning. Ian Goodfellow, Yoshua Bengio, Aaron Courville. MIT Press, 2016.
Theoretical and practical classes.
First epoch: deep learning project.
Second epoch: deep learning project.
Introduction to deep learning
History and cognitive basis of neural computation.
The perceptron / multi-layer perceptron
The neural net as a universal approximator
Optimization by gradient descent
Overfitting and regularization (Dropout)
Convolutional Neural Networks (CNNs)
Training with shared parameters: the convlutional model
Recurrent Neural Networks (RNNs)
Exploding/vanishing gradients Long Short-Term Memory Units (LSTMs)