Below are my recommendation on Learning Resources for ML and DL newbies. If you find yourself in the list, that means your work/material is excellent and helped me a lot.
Machine Learning Andrew Ng 吴恩达 ⭐️⭐️⭐️⭐️⭐️
ML is no longer rocket science🚀. It's now computer science💻.
Neural Networks for Machine Learning Geoffrey Hinton ⭐️⭐️⭐️⭐️
You will (and a lot of people did) get inspiration from him.
Andrej Karpathy's blog and his Course Materials for Stanford students. ⭐️⭐️⭐️⭐️⭐️
You will love him to the heart if you are a software hacker.
http://rll.berkeley.edu/deeprlcourse/ UCB ⭐️⭐️⭐️⭐️
Brand new course based on brand new (and soon to be popular) papers.
http://ai.berkeley.edu/lecture_videos.html UCB ⭐️⭐️⭐️⭐️⭐️
Traditional AI approaches to problems, with comics and humor. People interested in Deep RL could borrow a lot from Dan Klein's class.
David Silver's lecture and John Schulman's lecture exercises (both on YouTube) ⭐️⭐️⭐️⭐️
David Silver: Grandmaster Deep RL
John Schulman: lectures so-so, but the exercises are great
How to train a GAN series of videos and papers from Ian Goodfellow and friends. ⭐️⭐️⭐️⭐️
Actual tricks may vary but the intuitions are gold.
Paper on various topics. ⭐️⭐️⭐️⭐️⭐️
These papers discuss mainly about the possibilities and impossibilities of neural networks. Basically if you don't know how to (or whether you can) train a human for a task, you won't be able to train neural networks for that task. They explain exactly why.
Evolving Culture vs Local Minima Yoshua Bengio
Knowledge Matters: Importance of Prior Information for Optimization Çağlar Gülçehre, Yoshua Bengio
Reinforcement learning in the brain Yael Niv
Below is an excellent paper on human visual attention mechanism. If you know about human foveal receptive field patterns, you will find this work amazing.
EMERGENCE OF FOVEAL IMAGE SAMPLING FROM LEARNING TO ATTEND IN VISUAL SCENES Brian Cheung, Eric Weiss, Bruno Olshausen
Below: How to descent elegantly.
ADAM: A METHOD FOR STOCHASTIC OPTIMIZATION Diederik P. Kingma, Jimmy Lei Ba
Below: How to correctly initialize your network.
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification Kaiming He et al.