1/ DL 2.0: A Topological Tale of Everything wrong with Deep Learning
Not the Metrics We Want, But the Metrics We Need
Authors: Taha Bouhsine
Authors: Taha Bouhsine
The Google Developer Expert program and Google AI/ML Developer Programs team supported this work by providing Google Cloud Credit. I want to extend my gratitude to the staff at High Grounds Coffee Roasters for their excellent coffee and peaceful atmosphere. I would also like to thank Dr. Andrew Ng for creating the Deep Learning course that introduced me to this field, without his efforts to democratize access to knowledge, this work would not have been possible. Additionally, I want to express my appreciation to all the communities I have been part of, especially MLNomads, Google Developers, and MLCollective communities
If you use this work in your research, please cite:
GitHub Repo
Blog Post
Sponsor Project
Kaggle Notebook
DL 2.0 Paper
@article{bouhsine2024deep,
title={Deep Learning 2.0: Artificial Neurons That Matter - Reject Correlation, Embrace Orthogonality},
author={Bouhsine, Taha},
journal={arXiv preprint arXiv:2411.08085},
year={2024}
}
Bouhsine, T. (2024). Deep Learning 2.0: Artificial Neurons That Matter - Reject Correlation, Embrace Orthogonality. arXiv preprint arXiv:2024.xxxxx.
The source code, algorithms, and all contributions presented in this work are licensed under the GNU Affero General Public License (AGPL) v3.0. This license ensures that any use, modification, or distribution of the code and any adaptations or applications of the underlying models and methods must be made publicly available under the same license. This applies whether the work is used for personal, academic, or commercial purposes, including services provided over a network.