Course curriculum
-
1
Introduction
- Introduction to Instructor
- Introduction to Course
-
2
Motivation & Applications
- What Is Reinforcement Learning
- WhatIs Reinforcement Learning hiders and seekers by OpenAI
- RL vs Other ML Frameworks
- Why RL
- Examples Of RL
- Limitations Of RL
- Exercises
-
3
Terminologies of RL
- Introduction
- Envionment
- Agent
- Action
- State
- Goal and Done State
- Reward
- Fun Activity
- Policy and Plan
- Episode
-
4
Naïve Random Solution
- Introduction to Module
- Introduction to Game
- Rules of Game
- Setting up game Python pt 1
- Setting up game Python pt 2
- Setting up game Python pt 3
- Playing the game manually
- Implementing Random solution
- Q Learning and Q Table Theory
- Implemeting Q Learning pt 1
- Dry Run of get state
- Answer to Question
- Implemeting Q Learning pt 2
- Implemeting Q Learning pt 3
- Conclusion
-
5
RL based Q Learning Solution
- Introduction to Gym
- Frozen Lake Rules
- Implementing Frozen Lake pt 1
- Implementing Frozen Lake pt 2
- Implementing Frozen Lake pt 3
- Implementing Frozen Lake pt 4
- Agent plays the game
- Conclusion
-
6
Hyper Parameters & Concepts
- Introduction to Module
- Epsilon
- Updating Epsilon Value
- Gamma, Discount Factor
- Alpha Learning Rate
- Q Learning Equation
- Quiz (Number of Episodes)
- Solution (Number of Episodes)
- Quiz (Alpha)
- Solution (Alpha)
-
7
SARSA
- Introduction to SARSA
- Off policy VS On policy
- SARSA Implementation
- SARSA Implementation update
- Pros & Cons
-
8
DNN Foundation for Deep RL
- Why Deep Learning
- Why PyTorch
- PyTorch installation and Tensors intro
- Automatic Diffrenciation Pytorch New
- Why DNNs in Machine Learning
- Representational Power and Data Utilization Capacity of DNN
- Perceptron
- Perceptron Exercise
- Perceptron Exercise Solution
- Perceptron Implementation
- DNN Architecture
- DNN Architecture Exercise
- DNN Architecture Exercise Solution
- DNN ForwardStep Implementation
- DNN Why Activation Function is Required
- DNN Why Activation Function is Required Exercise
- DNN Why Activation Function is Required Exercise Solution
- MDP
- DNN Properties of Activation Function
- DNN Activation Functions in Pytorch
- DNN What is Loss Function
- DNN What is Loss Function Exercise
- DNN What is Loss Function Exercise Solution
- DNN What is Loss Function Exercise 2
- DNN What is Loss Function Exercise 2 Solution
- DNN Loss Function in Pytorch
- DNN Gradient Descent
- DNN Gradient Descent Exercise
- DNN Gradient Descent Exercise Solution
- DNN Gradient Descent Implementation
- DNN Gradient Descent Stochastic Batch Minibatch
- DNN Gradient Descent Summary
- DNN Implemenation Gradient Step
- DNN Implemenation Stochastic Gradient Descent
- DNN Implemenation Batch Gradient Descent
- DNN Implemenation Minibatch Gradient Descent
- DNN Implemenation in PyTorch
- DNN Weights Initializations
- DNN Learning Rate
- DNN Batch Normalization
- DNN Batch Normalization Implementation
- DNN Optimizations
- DNN Dropout
- DNN Dropout in PyTorch
- DNN Early Stopping
- DNN Hyperparameters
- DNN Pytorch CIFAR10 Example
-
9
Deep RL DQN
- Introduction & Recap
- DQN Algorithm Steps
- Introduction to Project (Cart pole)
- Policy Network Explained
- Neural Network Class Implementation
- Replay Memory & Experience
- Experience Implementation
- Replay Memory Implementatiton
- Target Network & Recap
- Epsilon Greeady Strategy Implemented
- Agent Class Implemented
- Environment Manager Implementation
- How to Get State
- Screen Preprocessing
- Screen Croping
- Screen Transformation
- Processed VS NonProcessed Screen
- Moving Avg Implemented
- Ploting the Moving Avg
- Hyperparameter Initialization
- Initializing the Classes
- Final Structure Implementation part 1
- Extracting Tensors
- Final Structure Implementation part 2
- Qvalues Calculator Implemented
- Removing Errors Final Structure Implementation part 3
- Visualizing the Training
-
10
StableBaseLines Cartpole Solution
- Introduction to Stable Baseline
- Loading & Understanding the Envireonment
- Train RL Model
- Evaluation and Testing
- Callbacks & Early Stopping
- Changing Policy Architecture
- Changing the Algorithm
- Tips for Accuracy Improvement
-
11
Trading Bot RL
- Introduction to Libraries and Project
- Loading the Data
- Setting up Environment
- Random Actions
- Training and Evaluating Model
-
12
Car Racing Game
- Introduction to Game
- Importing the Dependencies
- Exploring the Environment
- Training and Testing the Model
-
13
Interview Prep
- Prep 1
- Prep 2