/DeepLearningCourse

Code for my Deep Learning Course at University of Latvia

Primary LanguageJupyter Notebook

DeepLearningCourse

Code for my Deep Learning Course at University of Latvia

The course videos can be found here:

  1. Intro to Deep Learning: https://youtu.be/2K2RMy-5ZsM
  2. Bigram Model: https://youtu.be/Kbc-8iEe6Js (Original: https://youtu.be/kCc8FmEb1nY )
  3. Imitation learning (Data is King): https://youtu.be/8t4sxp7NVjg
  4. Numerical Optimisation: https://youtu.be/u-Sn0f2XHnI
  5. Architectures of Neural Networks: https://youtu.be/y6emEiI1lZw
  6. Intro to ChatGPT: https://youtu.be/HY-1RoVFftc (Course on prompt engineering ; GPT3 explained )
  7. Intro to Robotics: https://youtu.be/Bi9QtQb3vrA (Image segmentation example: https://github.com/paulispaulis/CLIC-semseg ; KarpathyAtTesla )
  8. From Bigram to ChatGPT: https://youtu.be/lZM36LLXWbc (Tokenization: https://youtu.be/zduSFxRajkE ; Attention: https://youtu.be/eMlx5fFNoYc ; What Transformers can learn: https://arxiv.org/abs/2310.16028)
  9. What is Language? https://youtu.be/aJ-3botLBM0 (Diversity of modern languages implies they emerged AFTER HomoSapiens migration out of Africa https://youtu.be/rQv7NBGsldk ; Impossible languages)
  10. Compression by LLM: https://youtu.be/r-fSI7rSjYI (Original: https://youtu.be/zjkBMFhNj_g )
  11. Evolutionary Learning: https://youtu.be/4OTuG0WRjaU (Demo https://youtu.be/GOFws_hhZs8 )
  12. Neural Network Basics: https://youtu.be/5KNE_zvnNeA
  13. Neural Network training (Backpropagation): https://youtu.be/GSZnx7VK3Zg (DetailedExplanation: https://youtu.be/VMj-3S1tku0 )
  14. PyTorch library: https://youtu.be/NlzC9zTXTHM
  15. Innovation: https://youtu.be/1Vhh_dYOu0I (Mystery2: Agentive evolutionary learning needs no training data)
  16. Technological Evolution https://youtu.be/wIJnvCqKPXM (Mystery1: Incremental 1-shot memory for continuous learning)
  17. Egocentric Viewpoint https://youtu.be/t3-InC9VEko (Mystery3: Concept creation)

Transcripts for all videos available from YouTube

In video description (press "...more" to reveal it all), at the lower-left corner there is a button "Show transcrpt". Copy/paste transcript to your favourite editor (remove timestamps with "Toggle timestamps" under symbol ⋮ in the top-right corner).

How these videos were voiced and transcribed?

Videos were subtitled and voiced using SELMA UC0 OpenSource software available at: https://github.com/SELMA-project/UC0-OpenSource