Colby Wang is an accomplished AI Engineer at Websters, bringing a wealth of experience in reinforcement learning, bandit problems, and generative adversarial networks (GANs). Colby’s journey in AI began with an intensive research internship under Professor Pascal Poupart at the University of Waterloo, where he specialized in reinforcement learning and bandits. During this time, he developed a novel approach to solve the Multi-Armed Bandit problem using Equivariant Neural Networks (ENN), achieving significant improvements in action probability learning, exploration, and exploitation.
In his role as an Undergraduate Research Assistant at the University of Waterloo’s Cheriton School of Computer Science, Colby implemented ENN models that demonstrated robust performance across diverse bandit environments. His work included achieving lower regret curves for Upper Confidence Bound and Thompson Sampling, fine-tuning models for diverse applications, and leveraging data consistency to optimize decision-making processes. These advancements in contextual bandits and reinforcement learning highlight his ability to innovate and solve complex AI problems.
Colby’s research experience extends to machine learning and GANs, where he collaborated with PhD student Kira Selby under the guidance of Professor Pascal Poupart. He contributed to the enhancement of SetGAN for high-resolution image generation, pretraining and optimizing the StyleGAN-XL decoder for ImageNet, and developing auxiliary networks to improve image label predictions. His work on integrating and debugging multiple codebases for the StyleGAN3-Editing network provided him with deep insights into model selection, evaluation, and project management under computational constraints.
In addition to his research roles, Colby has developed practical software applications, including a tool to draw catenary curves on PDFs and a website with adjustable features for handcrafted violin makers. This software has been used by artisans to enhance their craftsmanship, showcasing Colby’s ability to apply AI solutions to real-world problems.
Colby Wang’s academic and professional journey reflects his dedication to advancing AI technologies and his commitment to continuous learning and innovation. His expertise in reinforcement learning, bandit problems, and GANs makes him a valuable asset to the Websters team, driving forward the company’s AI capabilities and delivering cutting-edge solutions.
Colby Wang’s expertise and innovative approach continue to drive Websters towards new heights in AI engineering, ensuring the company remains at the forefront of technological advancements in the digital landscape.
“To be a CEO is a calling. You should not do it because it is a job. It is a calling, and you have got to be involved in it with your head and heart.”