Title: Automated Game Testing in the Era of Data-Driven Intelligence: Challenges and Opportunities
Abstract: Game testing has been long recognized as a notoriously challenging task, which mainly relies on manual playing and scripting-based testing in the game industry. Until recently, manually crafting test scripts still plays a dominant role in many game companies, which however could not satisfy the increasing quality demands of requirements as a software system in this speed era. The recent fast progress in data-driven AI (e.g., deep reinforcement learning, LLM) opens new opportunities to empower the various stages of game testing process with intelligence. In this talk, I provide a high-level introduction to our ongoing activities in automated video game testing empowered by data-driven AI. I will also highlight the challenges, opportunities, and possible future directions.
Bio:
Lei Ma is currently an Associate Professor with The University of Tokyo, as well as University of Alberta. He is also a Canada CIFAR AI Chair and Fellow at Alberta Machine Intelligence Institute (Amii). His research centers around the interdisciplinary fields of human-centered trustworthy software engineering (SE) and artificial intelligence (AI) with a special focus on the quality, reliability, safety and security assurance, as well as the interpretation and human interactivity of machine learning and AI Systems. Many of his works were published in top-tier software engineering, AI, and security venues (e.g., TSE, TOSEM, EMSE, ICSE, FSE, ASE, ISSTA, CAV, TNNLS, ICML, NeurIPS, ACM MM, AAAI, IJCAI, TDSC). He has received over 10 prestigious academic awards, including three ACM SIGSOFT Distinguished Paper Awards. For more detailed information, please visit the website, https://www.malei.org