Course Information
Registration period
Year-round Recruitment
Course Level
Study Mode
Duration
12 Hour(s) 24 Minute(s)
Language
English
Taught by
Jan Warchocki, SuperDataScience Team, Ligency
Certificate
- Available
- *The delivery and distribution of the certificate are subject to the policies and arrangements of the course provider.
Rating
Course Overview
Learn how to use powerful Deep Reinforcement Learning and Artificial Intelligence tools on examples of AI simple games!
Ever wish you could harness the power of Deep Learning and Machine Learning to craft intelligent bots built for gaming?
If you’re looking for a creative way to dive into Artificial Intelligence, then ‘Artificial Intelligence for Simple Games’ is your key to building lasting knowledge.
Learn and test your AI knowledge of fundamental DL and ML algorithms using the fun and flexible environment of simple games such as Snake, the Travelling Salesman problem, mazes and more.
1. Whether you’re an absolute beginner or seasoned Machine Learning expert, this course provides a solid foundation of the basic and advanced concepts you need to build AI within a gaming environment and beyond.
2. Key algorithms and concepts covered in this course include: Genetic Algorithms, Q-Learning, Deep Q-Learning with both Artificial Neural Networks and Convolutional Neural Networks.
3. Dive into SuperDataScience’s much-loved, interactive learning environment designed to build knowledge and intuition gradually with practical, yet challenging case studies.
4. Code flexibility means that students will be able to experiment with different game scenarios and easily apply their learning to business problems outside of the gaming industry.
‘AI for Simple Games’ Curriculum
Section #1 — Dive into Genetic Algorithms by applying the famous Travelling Salesman Problem to an intergalactic game. The challenge will be to build a spaceship that travels across all planets in the shortest time possible!
Section #2 — Learn the foundations of the model-free reinforcement learning algorithm, Q-Learning. Develop intuition and visualization skills, and try your hand at building a custom maze and design an AI able to find its way out.
Section #3 — Go deep with Deep Q-Learning. Explore the fantastic world of Neural Networks using the OpenAI Gym development environment and learn how to build AIs for many other simple games!
Section #4 — Finish off the course by building your very own version of the classic game, Snake! Here you’ll utilize Convolutional Neural Networks by building an AI that mimics the same behavior we see when playing Snake.
Course Content
- 13 section(s)
- 120 lecture(s)
- Section 1 Installation
- Section 2 Get the materials
- Section 3 Genetic Algorithms Intuition
- Section 4 Genetic Algorithms Practical
- Section 5 Q-Learning
- Section 6 Q-Learning Practical
- Section 7 Deep Q-Learning with ANNs
- Section 8 Deep Q-Learning Practical
- Section 9 Deep Convolutional Q-Learning
- Section 10 Deep Convolutional Q-Learning Practical
- Section 11 ANNEX 1: Artificial Neural Networks
- Section 12 ANNEX 2: Convolutional Neural Networks
- Section 13 Congratulations!! Don't forget your Prize :)
What You’ll Learn
- SOLVE THE TRAVELLING SALESMAN PROBLEM
- Understand and implement Genetic Algorithms
- Get the general AI framework
- Understand how to use this tool to your own projects
- SOLVE A COMPLEX MAZE
- Understand and implement Q-Learning
- Get the right Q-Learning intuition
- Understand how to use this tool to your own projects
- SOLVE MOUNTAIN CAR FROM OPENAI GYM
- Understand and implement Deep Q-Learning
- Build Artificial Neural Networks with Keras
- Use the environments provided in OpenAI Gym
- Understand how to use this tool to your own projects
- SOLVE SNAKE
- Understand and implement Deep Convolutional Q-Learning
- Build Convolutional Neural Networks with Keras
- Understand how to use this tool to your own projects
Reviews
-
LLuis Ramirez
excelente elección
-
LLeandro Westin
Very didactic and explanatory!
-
YYesenia Monterroso
me gusta mucho lo de la programacion me esta costando entender espero poder aprender un poco.
-
AArtem S.
Q-Learning was explained in mathematical language -- this is the best explanation I can give. It was 2 hours of math-heavy theory, where the problem was dissected into many smaller ones, and the smaller problems were explained from the mathematical standpoint, but the bigger picture never came together in my head. No practical example of a Q-Learning algorithm was given from the programming point of view - important parts were omitted entirely from the explanation. In the end, you're left with a "magic formula", which if you don't understand, that's it for you - you won't grasp Q-Learning. The theory was followed by an hour long writing of an algorithm that's just 5 lines long at its core, so that's an hour wasted. But the 5 lines can be summed up as "remember the math formulas? well here they are", again without the explanation of their practical purpose. I had to "step into" the algorithm in my head to figure out what it was doing step by step, and I understood in the end, but I didn't understand the whys of it all. I had to talk to a friend knowledgeable in AI to understand what the formula is trying to do and why. Now that I understand it myself, I could explain it with no math involved and in under 5 minutes to anyone. Instead, the videos went for a purely academic approach, with little regard to practicality. I can't rate this course very high. I may update the review after watching other sections, but so far it's been a very frustrating experience.
