[Deep Learning Matrix Reloaded] From spanning sets to deep matrix factorization (part 2)

#IntuitiveDeepLearning #SimpleMathematicsOfDL #linearalgebra 📚🔗 The summary map of the DL linear algebra of the first lecture videos 1.1 to 1.6 can be found at: https://drive.google.com/file/d/198s8Tr8AGHplQaRZXcLawshiakGTxXKA/view?usp=sharing Unlock the world of Deep Learning with our new “Intuitive Deep Learning” video series! 🚀 Dive into the mathematical foundations of DL using an intuitive approach and the power of linear algebra. 🧠 We hope that you will find the joy of clarity and multi-layer understanding of DL methods in this new series. In this DL map video we briefly cover the following topics: 1. Perfect data representation using a fixed spanning set 2. Perfect data representation using an orthonormal spanning set (linear autoencoder) 3. Imperfect data representation using a fixed spanning set 4. PCA or linear autoencoder formalized using matrix factorization 5. K-mearns using matrix factorization 6. Sparse coding using matrix factorization 7. Generalized matrix factorization 📊💡 Join us on this journey to gain a simplified understanding of complex concepts in DL. Special thanks go to Jeremey Watts, Simon Prince and co-authors for producing foundational textbooks: • Machine Learning Refined; https://github.com/jermwatt/machine_learning_refined • Understanding Deep Learning; https://github.com/udlbook/