Goal
Upon successful completion of the course, the student will be able to:
- Explain fundamental concepts and principles regarding deep machine learning and the parameterization/implementation/evaluation of deep machine learning algorithms on datasets.
- Know, configure and apply the most basic deep machine learning algorithms per problem type category (regression, classification, clustering, policy optimization).
- Know the methodology of applying and evaluating deep machine learning algorithms to data, comparing and choosing the appropriate algorithm.
- Understand algorithms, select, design or adapt the most appropriate and apply/evaluate to areas of interest
- Communicate deep machine learning ideas in a clear, concise and formal manner. Aiming to parameterize, apply and evaluate deep learning algorithms to problems, and explain their operation.
Also, the course targets to the following general competencies:
- Ability to organize and plan work, and manage time effectively
- Ability to communicate effectively (orally and written)
- Ability to solve problems
- Ability to develop critical thinking and capacity for critical approaches
- Ability to work in a team
- Ability of interdisciplinary approaches
- Ability to apply theoretical knowledge in practice
- Ability to research
- Ability to adapt methods and techniques to new situations and conditions
- Ability to generate new ideas – Creativity
Contents
- An introduction to deep learning
- Feed-Forward Neural Networks
- Convolutional Neural Networks
- Autoencoders and Data Augmentation
- Recurrent and Recursive Neural Networks
- GRU/LSTMs & Attention 21 2024-2025
- Deep Reinforcement learning: Introduction and algorithms
- Imitation learning
- Policy gradient, value-based & actor-critic algorithms
Case studies, example problems and methods for solving them are presented.
Bibliography
- “Deep Learning”, Ian Goodfellow and Yoshua Bengio and Aaron Courville, MIT Press,
2016.
- “Reinforcement Learning: An Introduction”, Richard S. Sutton and Andrew G. Barto Second Edition (see here for the first edition) MIT Press, Cambridge, MA, 2018
- Additional research articles on the topics of the lectures.
