Neural Networks and Deep Learning

*** The first lecture is scheduled on January 11, 2022 at 9:00.
All lectures will be in presence in the GRAY ROOM of the TeCIP Institute, on each Monday and Tuesday, from 9:00 to 11:30.


To attend the lectures and receive notifications about any change on course lectures, please register in this file.

This course (6 CFUs) includes two modules of 3 CFUs each: the first module focuses on the theoretical foundations of neural networks and deep learning, while the second module covers more practical and implementation issues.

Course Program

Course topics: theoretical part

  1. Introduction to neural computing
  2. Hopfield networks
  3. Unsupervised learning
  4. Reinforcement learning
  5. Supervised learning - part 1
  6. Supervised learning - part 2
  7. Towards Deep Neural Networks (DNNs)
  8. Autoencoders
  9. Convolutional Neural Networks
  10. DNNs for object classification
  11. DNNs for object detection
  12. Recurrent Neural Networks
  13. Deep Reinforcement Learning
  14. Generative Adversarial Networks
  15. Sample applications and open issues

Slides on the practical part

  1. Implementing Hopfield networks in C
  2. Implementing Kohonen networks in C
  3. Implementing Reinforcement Learning in C
  4. Implementing Backpropagation in C
  5. Frameworks for training and inference of DNNs
  6. Modeling DNNs in Tensorflow and Caffe
  7. Overview of the Apollo framework for Autonomous Driving
  8. Networks in Apollo: Perception Module
  9. Neural-based control: OpenAI Gym simulation environments
  10. Deep Reinforcement Learning: DDPG
  11. Genetic Algorithms for control by Reinforcement Learning
  12. Adversarial examples: attacks and defense techniques
  13. Implementing DNNs on GPUs platforms - Part I
  14. Implementing DNNs on GPUs platforms - Part II
  15. Accelerating DNNs on FPGA platforms
  16. The CHaiDNN framework
  17. The Vitis AI framework