I’ve completed several courses from DeepLearning and Coursera for AIDL using Python.
One of the most interesting topics is Multi-head masked self-attention in Transformers. That’s got to be one of the most remarkable breakthroughs in Computer Science of our time. Then there’s back propagation at core of Neural Networks which is optimizing a cost function using derivatives from Calculus. If you think about it, that is really incredible also. IIRC backprop was invented/discovered in the 1950’s or so. I wonder what those researchers would think if they were alive today.
Natural Language Processing with Attention Models
https://www.coursera.org/account/accomplishments/certificate/39R93C6PG2K3
Natural Language Processing with Sequence Models
https://www.coursera.org/account/accomplishments/certificate/FJTVCXJ2EZMQ
Natural Language Processing with Probabilistic Models
https://www.coursera.org/account/accomplishments/certificate/NQB2RM5MQR2R
Natural Language Processing with Classification and Vector Spaces
https://www.coursera.org/account/accomplishments/certificate/CNAT67ZYNBA2
Sequence Models
https://www.coursera.org/account/accomplishments/certificate/22FN5NM7SZ6Q
Convolutional Neural Networks
https://www.coursera.org/account/accomplishments/certificate/BTG9PEYCXT6D
Structuring Machine Learning Projects
https://www.coursera.org/account/accomplishments/certificate/EMU4U7MVBPK8
Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization
https://www.coursera.org/account/accomplishments/certificate/JDNU4ELTQ4Z5
Neural Networks and Deep Learning
https://www.coursera.org/account/accomplishments/certificate/5ESUBYG4LKYD