Good Resources

Collections of resources (for machine learning, math, statistics) that I wish I could have known them earlier…
not a comprehensive list, please let me know (email or Twitter) if you want to add any!

Courses

• Learning from data: introductory Machine Learning online course, taught by Caltech Professor Yaser Abu-Mostafa (one of the most amazing lecturers I’ve ever met!). A learning theory-oriented machine learning course.
• Stanford cs231n: CNN: introduction for deep learning (CNN, RNN, CV oriented)
• Deep Learning NYU - Yann LeCun & Alfredo Canziani: course run in 2020, website/notes here
• Statistics from Professor Leonard: the courses are quite easy to follow and interesting! Learn some basic statistics concepts there.
• Bioinformatics by Pavel Pevzner

Tutorials

• 3Blue1Brown: Excellent short tutorials with vivid illustration and clear explanation. Series I found interesting and useful: COVID19, Essence of linear algebra, Essence of calculus
• The Bright Side of Mathematics: very clear and intuitive! I learn Measure Theory there!

Gaussian Process

• Deep Gaussian Process: Neil D. Lawrence - NIPS Tutorial 2017

Knowledge Graph

• Stanford CS520 Knowledge Graphs Seminar (Spring 2020)

Fourier Transform

• (Chinese material) 李永乐老师: 傅里叶变换
• 3Blue1Brown:
  • what is Fourier Transform
  • Fourier Series
  • Euler’s formula with introductory group theory
• Reducible: FFT

Deep Learning related

• brief intro to Transformer in 10min: by CodeEmporium
• NLP_ability git repo (in Chinese): a comprehensive summary of NLP-related stuff, suitable for beginners. I know this from a good explanation to Transformer video.
• A tutorial for PyTorch by Aladdin Persson: very detailed and easy-to-follow. Highly recommended!

Summer Schools

• Machine Learning Summer School: I attended MLSS2020 and learnt a lots from the amazing speakers and also participators. Good places to make friends and discuss your research ideas!
• Gaussian Process Summer School

Books

• Pattern Recognition and Machine Learning, Foundations of machine learning: textbooks of machine learning
• Gaussian Process for Machine Learning
• Mathematics for Machine Learning: provides the necessary mathematical skills to read those other machine learning books. Highly recommended for beginners!
• Algebra, Topology, Differential Calculus, and Optimization Theory For Computer Science and Machine Learning: A very comprehensive and well-written book from Jean Gallier and Jocelyn Quaintance.
• Bandits Algorithms by Tor Lattimore and Csaba Szepesvari
• Introduction to Reinforcement Learning
• All of statistics

Websites/Blogs

• Simons Institute for Theory of Computing: good online programs/events upcoming!
• I’m a bandits-Sébastien Bubeck: Random topics in optimization, probability, and statistics.

Mentoring

• mentor.net: I went several mentoring sessions during NeurPIS2020, they were quite good. There are also some from time to time.

My favorite TED talks

• Uri Alon: Why truly innovative science demands a leap into the unknown: Being supportive and say: Yes! And?
• The first 20 hours – How to learn anything: You do not need 10-thousand hours to LEARN something, but only 20 hours (well, you do need more time to become an expert).
• How language shapes the way we think: “To have a second language is to have a second soul” -Charlemagne.