Advanced games AI with genetic algorithms, neural networks & Q-learning in C# and Tensorflow for Unity
InstructorPenny de BylThe structure moves logically from genetic algorithms to individual perceptrons, then to full artificial neural networks and reinforcement learning.
The curriculum includes various challenges such as camouflage training, maze walking, and pong gameplay to demonstrate machine learning in action.
Editorial course preview
These 4 complementary views highlight concrete, legible examples from the course presentation.
Unity Editor split-screen view displaying a genetic algorithm simulation with humanoid agents on a red platform, showing scene perspective and game view with population statistics.
This diagram illustrates the architecture of a neural network, featuring an input layer, two hidden layers, and an output layer to visualize data flow.
This image displays the Unity Editor environment featuring a Pong game simulation alongside the corresponding C# neural network code used to train the AI agent.
This educational slide defines Q-learning by presenting the formula for the quality of performing an action alongside Bellman's equation for calculating maximum future rewards.
One sampled review suggests that the instructor is noted for her ability to simplify complex mathematical and machine learning concepts for a wide range of learners.
The curriculum contains specific sections dedicated to older, deprecated versions of Unity's ML-Agents (V0.2 and V0.3).
Learners have reported that significant changes in Unity since the course release can cause broken workflows and require deviating from the lessons.
One sampled review suggests that some learners have noted that presentation slides can be difficult to read or lack legibility.
The curriculum aligns with the target audience of Unity developers interested in machine learning algorithms.
Free