The course is organized in three-hour sessions in the morning and two-hour sessions in the afternoon. This schedule leaves ample time to assimilate the material, run codes and exchange experience with classmates and the instructor.

Day 1. Introduction to Supervised and Unsupervised Machine Learning

Wednesday, 13 January 2021

  • What is Machine Learning
  • Supervised vs Unsupervised Learning
  • Linear Regression
  • Logistic Regression
  • K-Means
  • Principal Component Analysis.

Day 2. Feed-Forward Neural Networks

Thursday, 14 January 2021

  • From Logistic Regression to Single Neuron Representation
  • Feed-Forward Neural Networks
  • Back-Propagation
  • Batch, Stochastic and Mini-Batch Gradient Descent

Day 3. Bias and Variance in Neural Networks

Friday, 15 January 2021

  • Overfitting and Underfitting
  • Regularization
  • Batch Normalization
  • Machine-Learning Diagnostics
  • Training Set, Validation Set, Test Set and k-Fold Validation

Day 4. Convolutional Neural Networks

Monday, 17 January 2021

  • Convolution Layers: Definition and Notations
  • Pooling and Fully Connected Layers
  • Transfer Learning
  • Large Datasets and Well-Known Convolutional Neural Networks

Day 5. Recurrent Neural Networks

Tuesday, 18 January 2021

  • Recurrent Networks
  • Long Short-Term Memory (LSTMs)
  • Moving towards Generative Networks: Probabilities for Machine Learning
  • Large Datasets and Well-Known Convolutional Neural Networks

Day 6. Generative Neural Networks

Wednesday, 19 January 2021

  • Autoencoders
  • Variational Autoencoders
  • Generative Adversarial Networks and some of their Applications in Geoscience

Deep Learning

Deep learning is broad family of machine learning methods based on artificial neural networks. Applications are extremely diverse.


The course is full. You can still register on a waiting list:

Registration form.