4 books - terrytaylorbonn/auxdrone GitHub Wiki

26.0104 [email protected], linkedin.com/in/terry-taylor-biz, Lab notes (Gdrive), Git

Available for download from my gdrive.

__ziptieai_BOOK1_CNN_

I was doing demos and deploying on drones a few years ago. These are mostly pic recognition, etc. CNNs could be considered to be forerunner of LLM/TF and JEPA.

  • AlexNet (a milestone AI CNN in 2012) analysis. For now this book is just about AlexNet.
  • Various drone demo AI. These are documented in other docx's on the gdrive. I might copy some of them into this book. o object detection, recognition, classification o segmentation, depth
  • maybe UFA (the theory) (universal function approximator). the theory, maybe some demos.

__ziptieai_BOOK2_LLM_ (previously _the_kentuckian_guide_to_ai_BOOK1_LLM)

  • My original plan was just to create this book.
  • LLMs and transformers (TFs).
  • Need to add ViT (video transformers)?
  • Youtube overview

Most recent work leading up to the book:

__ziptieai_BOOK3_JEPA_

  • Yan LeCun really sparked my interest in JEPA. It will be basically a world model for robots. JEPA basically builds on CNN and TFs.
  • I-JEPA (images)
  • V-JEPA (video)  

__ziptieai_BOOK4_GI_

For demos (first on 26.0103) see https://drive.google.com/drive/folders/1-Adawag9uA8_bq-hDF-nOuPYaRLz1eEO #500_jepa_.docx

This doc 4 will be a "clean break" from the previous 3. Its about electronic analog 3d (non-GPU, non-CPU, non-2d) that are not based on "crude" binary algorithms of brain function, but attempt to replicate the complex symphony of signals in the brain (its not just the wires that carry electrical signals, but also the EM crosstalk and interplay between these wires).

  • For now this is just my own theory (I arrived at independently; surely someone else has written about this).
  • Mainly study of biological structures and
  • how these could be replicated with 3d electronic structures
  • resulting in genuine intelligence (GI) (real consciousness, life that operates at light speed)