Where to start RL?
Hi, I am a developer and have always been interested in ML and especially in RL. I finally want to start learning. I have a basic understanding of ML and training.
From my understanding, I should start with revising basic maths and trying some basic coding projects before going deep. Please suggest what I should cover and any courses that I can look at. Deep Reinforcement Learning Course by Huggingface seems interesting.
I am not asking for zero-to-hero steps in a month (I know that's impossible); I am willing to spend time daily and give it a genuine try.
All the suggestions, advice and personal experience are welcome. Thanks in advance.
Update:
This is what I have organised so far. I am not promoting any channels. This is what I found online as good resources:
- [ ] Linear Algebra
- [ ] 3Blue1Brown Essence of Linear Algebra
- [ ] Matrix shapes
- [ ] dimensions
- [ ] Tensors - Coding in PyTorch
- [ ] Calculus Intuition
- [ ] 3Blue1Brown Calculus Series
- [ ] Partial Derivatives & The Chain Rule: https://www.youtube.com/watch?v=IN2XmBhILt4
- [ ] Matrix Calculus: https://www.youtube.com/watch?v=IN2XmBhILt4&t=93
- [ ] Probability
- [ ] Khan Academy Probability and Statistics
- [ ] probability basics
- [ ] conditional probability
- [ ] expected value
- [ ] Markov Property (Memorylessness): https://www.youtube.com/watch?v=ljrjEh13Vyg
- [ ] Bayes' Theorem: Watch 3Blue1Brown: Bayes' Theorem Visually
- [ ] Statistics Basics
- [ ] mean
- [ ] variance
- [ ] standard deviation
- [ ] Functions & Graphs
- [ ] what a function is
- [ ] input/output
- [ ] graph intuition
- [ ] slope intuition
- [ ] ML/RL Math Intuition
- [ ] StatQuest
- [ ] https://www.youtube.com/playlist?list=PLblh5JKOoLUIxGDQs4LFFD--41Vzf-ME1
- [] https://www.youtube.com/watch?v=NFo9v_yKQXA
- [ ] Reinforcement Learning: An Introduction by Andrew Barto and Richard Sutton