Into the Unknown
Bayesian Methods for Machine Learning (with Honors)Higher School of Economics This is an excellent course from Russia and almost only for those who love mathematics and enjoy the procedure of derivation. Topics range from basic Expectation-Maximization to advanced such as Variational Autoencoder, and Bayesian Optimization. Programming assignments cover a few powerful libraries – PyMC, GPy,…
Neural Networks for Machine LearningUniversity of Toronto Taught by ‘godfather’ Geoffrey Hinton, this 16-week course is definitely a masterpiece, which covers almost entire modern AI history ever since 1960 – tons of theories, equations, derivations, examples, and anecdotes. The topics like Boltzmann machine, belief nets, auto-encoder are rather tough, but they help me ‘re-build’ my…
Applied Machine Learning in PythonUniversity of Michigan This is a nice course about scikit-learn – a stable, powerful and still indispensable library nowadays. The course covers many classic models, and how to use them efficiently. Various evaluation curves and model selection are what I like most. The comparison between scikit-learn with other new libraries (such…
Structuring Machine Learning ProjectsDeepLearning.AI This is a short-but-effective course about strategy. The topics are all critical for helping team analyze errors, select direction, make progress and archive goals effectively. The methods demonstrated when doing analysis are very impressive. I am afraid these knowledge will never be available in any academic textbook, you can only learn…
Improving Deep Neural NetworksDeepLearning.AI This course guides you through the “mist” of optimizing deep learning neural networks, very clear and joyful. RMSprop / Adam algorithms, Batch Norm, everything is elaborated carefully. Final programming assignment is to use TensorFlow to create a neural network. Fantastic course! I am very grateful. My Certificate Related Specialization I am…
Probabilistic Graphical Models 2: Inference (with Honors)Stanford University A very challenging course! The programming assignments (clique tree, sampling methods) almost killed me, but gladly I made it. Happy New Year 2018, more challenges coming soon. My Certificate Related Specialization I am Kesler Zhu, thank you for visiting. Checkout all of my course reviews at http://KZHU.ai
Programming for Everybody (Getting Started with Python)University of Michigan Oh, I love this course, the lecturer Prof. Charles Russell Severance is really a fun guy, his good humor is infectious! This is really a beginner level course that I finished in a just a few hours. No, I am not boasting. I am saying this…
Probabilistic Graphical Models 1: Representation (with Honors)Stanford University HAHA, certificate with Honors! PGM, it is said this is one of the hard courses in Stanford, but totally worth it! All Programming Assignments are challenging (Bayesian net, Markov net/CRF, and decision making), but very essential to help understand how PGM works. I definitely will enroll the…
Neural Networks and Deep LearningDeepLearning.AI GREAT course! Andrew elaborated the neural network for Deep Learning in an extremely clear and well-organized way! 鉂わ笍 Follow his guidance and instructions, not only I comprehended the intricacies of deep learning, but also a full-fledged neural network can be implemented from scratch by myself in Python! My Certificate Related…
Introduction to Data Science in PythonUniversity of Michigan This is a nice course to get familiar with NumPy, Pandas, SciPy, etc. Lecture video is not difficult, but Programming Assignment is a little bit challenging and truly inspiring. My original code costs 10 minutes to manipulate a DataFrame with 10 thousand rows, but there is a…
Machine LearningStanford University The course is taught by Andrew Ng. Probably this is the most popular and famous MOOC course on the Internet. It is also the primary reason why I signed up and started my journey on Coursera. I spent almost 3 months to finish this course, I still remember how thrilled I was…