Lectures

Slides & recordings

Lecture plan:

• 16.10 MM - backprop
• 23.10 MM - optimizers
• 30.10 MM - helping SGD
• 06.11 MM - wrap up, MC - frameworks
• — bonus material - prerecorded lecture - practical considerations
• 13.11 MC Convnets 1 (classification)
• 20.11 MC Convnets 2 (detection & segmentation)
• 27.11 MC Generative modeling (VAE, GANs)
• 4.12 MC RNNs
• 11.12 MC Language Modelling
• 18.12 ? Transformers
• — bonus material (if needed)
• 8.01 MC RL1 - DQN
• 15.01 MC RL2 - Policy Gradients
• 22.01 MC RL3 - AlphaGo, ChatGPT

Labs

• Lab1:
  • lab1-cross-validation
  • lab1-MSLE
• Lab2: lab2-backprop
• Lab3: lab3-optimizers
• Lab4:
  • pytorch-tutorial
  • lab4-mnist
• Lab5: lab5-batchnorm-and-convnets
• Lab6:
  • goldfish-to-shark
  • visualization-by-occlusion
• Lab7:
  • residual-block
  • unet
• Lab8:
  • CGAN
• Lab9:
  • VAE
• Lab10:
  • Transformer
  • RNN

Homeworks

• Homework 1
• Homework 2

Old exams

Both coding tasks and tests from editions 2020/21-2023/24 are here