Machine Learning

• PyTorch Get Started
  • Credit
  • WHAT IS PYTORCH?
  • AUTOGRAD
  • NEURAL NETWORKS
    • Kernel with padding
    • Stride (3,2) (height,width)
  • TRAINING A CLASSIFIER
• PyTorch Tutorial DQN
  • The Theory
  • The Algorithm
  • The PyTorch code
    • 1. imports
    • 2. using namedtuple
    • 3. the ReplayMemory class
    • 4.1 the neural network class without fully connected layer (finding parameter for fully connected layer)
    • 4.3 the neural network class: Image processor added
    • 5. the Agent class (the main algorithm as explained earlier)
    • 6. mygame class (this is totally up to you)
    • 7. the main loop
  • Code Rewrite
• PyTorch Tutorial RNN 1
• PyTorch Tutorial RNN 2
• Awesome Stock Ai
  • Recycle Bin
• PyTorch Tutorial Transformer
  • Transformer Model
  • Data And Training
• Reinforcement Learning Theory
  • Origin
  • Definition of RL
  • Bellman equation: A self-consistency equation just like in Density-functional theory
  • policy gradient
  • Soft Actor-Critic
  • Policy Optimization vs Q-Learning
  • Algorithms
    • Vanilla Policy Gradient
    • 2015 Trust Region Policy Optimization (not in comparisons)
    • 2017 Proximal Policy Optimization
    • 2016 Deep Deterministic Policy Gradient
    • 2018 Twin Delayed DDPG
    • 2018 Soft Actor-Critic