
Abstract: Nowadays, we can’t escape from seeing social media posts showing how AI creates amazing results. This includes using deepfake to swap face with actor in Hollywood films or transforming face into different gender or age. Indeed, any sufficiently advanced technology is indistinguishable from magic. This workshop will walk you through how to convert mathematical concepts into code to build AI models. At the end of the workshop, you’ll learn how to write code from scratch to do the magic i.e. generate images and deepfake video.
This workshop is for people who want to understand how generative AI works and especially for those who want to implement the models themselves. Knowledge of TensorFlow/PyTorch and convolutional neural networks is preferred to get the best out of this workshop.
Bio: Soon-Yau Cheong is the founder of Sooner.ai. It helps businesses to devise and implement AI strategies. Past projects include helping ARM to research using computer vision in automotive camera, and scaling up federated learning software infrastructure with Samsung Research. He has a wide interest in many AI domains including computer vision, NLP and productizing AI. Soon-Yau is also the author of “Hands-on Image Generation with TensorFlow” where he implemented many state-of-the-art models from scratch.

Soon Yau Cheong
Title
Founder | Author of “Hands-on Image Generation with TensorFlow” | Sooner.ai
