Deep Reinforcement Learning: Hands-on AI Tutorial in Python

Name: Deep Reinforcement Learning: Hands-on AI Tutorial in Python
Rating: 3.9 (225 reviews)

Develop Artificial Intelligence Applications using Reinforcement Learning in Python.

Created byMehdi Mohammadi

Last updated 10/2020

English

What you'll learn

The concepts and fundamentals of reinforcement learning
The main algorithms including Q-Learning, SARSA as well as Deep Q-Learning.
How to formulate a problem in the context of reinforcement learning and MDP.
Apply the learned techniques to some hands-on experiments and real world projects.
Develop artificial intelligence applications using reinforcement learning.

Coding Exercises

This course includes our updated coding exercises so you can practice your skills as you learn.

Course content

6 sections • 51 lectures • 4h 3m total length

Introduction1:44
Course Structure1:23
Environment Setup1:13

Markov Decision Process2:38
Optimal Policy6:47
Bellman Equation3:20
The Bellman equation defines q(s,a) as the immediate reward plus the discounted maximum q-value of the next state over all possible actions.
Q-Learning2:26
Step-by-step Example8:08
Sarsa2:10
Deep Q-Network5:51
Exploration vs. Exploitation2:53
Define RL Problem - Examples3:52
Define reinforcement learning problems by specifying environment, states, agent, actions, and a reward function; update Q values via the equation, illustrated by navigation grid and job scheduling.
Reinforcement learning algorithms
SARSA algorithm

Overall Design3:05
Maze-based reinforcement learning project on a 6x6 grid features a red circle agent that uses four directions to reach the goal, with rewards +1 for goal and -1 for holes.
Create Project0:53
Create a reinforcement learning project for a maze problem using a pie chart editor, choosing Q-learning, Python 3.6, and creating a folder in My Documents.
Create files1:01
Create Maze Environment class5:07
Implement Building Maze Grid13:36
Create a maze grid by building a canvas, drawing vertical and horizontal lines, defining the origin, and placing the start, walls, goal circle, and a red explorer rectangle.
Test build_maze method1:20
Test and debug the build_maze method by using the main block and main loop to run the application window, ensuring the function executes correctly and the program starts.
Render and Reset methods4:19
Implement getting next state and reward12:38
Create Agent class4:02
Implement adding states3:30
Implement choosing action4:27
Implement the choose action function with an epsilon-greedy strategy to balance exploration and exploitation, and select either a random action or the best state-action.
Implement learn method6:25
Create App2:51
Implement main method10:01
Implement plotting results3:52
Run the App4:45
Expand Maze Environment

Overall Design3:15
Build a stock trading agent with a deep Q network, using a sliding window of closing prices to define states and actions like buy, sell, or hold, aiming for profit.
Start project1:35
Prepare dataset0:34
Create Market Environment class2:37
Implement getting data3:05
Implement getting all states7:15
Implement getting next state and reward6:56
Create Agent class8:23
Implement creating deep learning model and reset method4:54
Implement getting action10:37
Implement buy and sell7:42
Implement experience replay8:26
Create training app16:06
Create training app guides implementing a reinforcement learning training loop by configuring window size and data, running episodes, learning from market interactions, and saving the trained model.
Test training app1:01
Create evaluation app8:14
Implement plotting results6:30
Run training and evaluation1:33
Extending Stock Trading with Multiple Features1:14
Multiple Feature Stock Trader6:55

Requirements

Students are assumed to be familiar with python and have some basic knowledge of statistics, and deep learning.

Description

In this course we learn the concepts and fundamentals of reinforcement learning, it's relation to artificial intelligence and machine learning, and how we can formulate a problem in the context of reinforcement learning and Markov Decision Process. We cover different fundamental algorithms including Q-Learning, SARSA as well as Deep Q-Learning. We present the whole implementation of two projects from scratch with Q-learning and Deep Q-Network.

Who this course is for:

Machine learning and AI enthusiasts and practitioners, data scientists, machine learning engineers.

Deep Reinforcement Learning: Hands-on AI Tutorial in Python

What you'll learn

Explore related topics

Coding Exercises

Course content

Introduction3 lectures • 4min

Jump into Reinforcement Learning3 lectures • 11min

RL Algorithms9 lectures • 38min

Hands-on Project 1 - Maze Problem16 lectures • 1hr 22min

Hands-on Project 2 - Stock Trading19 lectures • 1hr 47min

Summary1 lecture • 2min

Requirements

Description

Who this course is for: