The field of machine learning has advanced tremendously in the past few years, and canny game makers are constantly finding new and interesting ways of applying machine learning techniques to build better games.
At the 2019 Game Developers Conference in March and you'll have the chance to dig in and spend a full day learning from some of the best and brightest in the game industry at the brand new GDC 2019 Machine Learning Tutorial!
This is just one of many great Bootcamps and Tutorials scheduled during the first two days of GDC (Monday and Tuesday, March 18th and 19th this year), albeit one that offers an up-to-the-minute, laser-focused look at the art and business of making and running games that make smart use of machine learning techniques.
For example, in "Beating Wallhacks Using Deep Learning With Limited Resources" Nexon Korea machine learning engineer Junsik Hwang will show you how Nexon Korea has developed a real-time automated wallhack detection system using Convolutional Neural Networks with a small dataset and a single GPU.
By using Class Activation Maps, the network finds suspicious areas within a screenshot that improves the credibility of the model's performance and makes debugging datasets much more efficient. Model Interpretability plays a crucial role in incorporating deep learning with the existing abuser control policies. As a result, the system now detects abusers in real-time and reduces manual inspection labor significantly!
And in "Simple Head Pose Estimation for Dialogue Wheels", Remedy lead character technical artist Antti Herva will show you a machine learning project aimed at helping animators liven up dialogue wheel interactions for an upcoming Remedy game project..
Make time to catch this talk if you want an introduction on performance capture, selecting image features and machine learning models, annotating data, training a neural network and finally evaluating the results in-game!
Plus, Electronic Arts' Fabio Zinno will be presenting a Machine Learning Tutorial talk on "From Motion Matching to Motion Synthesis, and All the Hurdles In Between" that will give you an expert overview of state-of-the-art ML techniques (Phase-Functioned Neural Networks and Mode-Adaptive Neural Networks) that use neural networks to synthesize motion from examples. Zinno aims to explicitly call out important architecture and implementation details, and spark a discussion on how this technology can be used in a modern game development pipeline.
And you won't want to miss "Smart Bots for Better Games: Reinforcement Learning in Production", a presentation from Ubisoft data scientist Olivier Delalleau about various reinforcement learning algorithms and how they may help game studios create better games, more efficiently.
Besides AI development, the ability to train bots to play games during production opens up promising opportunities for automated testing and design assistance. But applying reinforcement learning to modern games brings up many challenges, illustrated through several examples, with a focus on recent experiments within Ubisoft games. Whether you want to directly learn from pixels to minimize the integration burden, or entirely rewrite your engine to make it more "reinforcement learning-friendly", this presentation is packed with practical tips to help you reach your goal without (too many) tears.
For more details on these and all other announced talks head over to the online GDC 2019 Session Scheduler, where you can plan out your week at the show.
Bring your team to GDC! Register a group of 10 or more and save 10 percent on conference passes. Learn more here.