Ben Weber - Computational Media Alumnus

Ben weber imageBen Weber received his PhD in Computer Science from UC Santa Cruz after attending from 2007 to 2012, where he focused on game artificial intelligence and data mining. His dissertation involved building a bot for the RPG StarCraft 1 that learned how to play the game from replays of professional players and combined this knowledge with hand-authored behaviors. During his time at UCSC, Weber started the StarCraft AI Competition, which is now an annual event co-organized with the AI and Interactive Digital Entertainment (AIIDE) conference. Since graduating, Weber has worked in analytics and data science roles at several gaming companies. In October, he started a new position at Twitch working as a data science manager. His personal website lists his experience, projects, blog and related press.

How and when did you decide to study computational media?
During my second quarter, I took the Game AI course with Michael Mateas. I came to UCSC wanting to do research in artificial intelligence, and until this course I was unaware of the opportunity to work with games as part of my graduate study. I was fascinated by the material, and the following quarter I continued my project as independent study and joined the Expressive Intelligence Studio.

What experience or course in computational media has helped you most in your career? 
The guest lectures and visiting gaming industry professionals were one of best parts of the program. The faculty has a great network and the proximity to Silicon Valley meant that students had several opportunities to attend talks, show off their research, and network with professionals in the gaming industry. I was able to get an internship at Electronic Arts because of the relationships that the faculty had established. 

EIS Bot videoWhat are your long-term professional or creative goals with regards to computational media?
I continue to be interested in building systems that can learn from human behavior to create more engaging interactive experiences. For my dissertation work, I built a bot that learned how to play by studying replays from professional players. More recently, I’ve done work with recommendation systems that identify novel content that players may enjoy. At Twitch, there’s a lot of opportunity to learn from our viewers and broadcastings to create more engaging experiences.  

Would you recommend the Computational Media program at Baskin School of Engineering?
Yes! There’s great faculty, the proximity to Silicon Valley, and a beautiful campus. Students are able to get hands-on experience with building games and interactive experiences, and there are opportunities for internships with companies in the bay area. There’s also the chance to interact with students in other departments, I attended a few guest lectures from the statistics department and played pickup soccer with many of these students.  

Ben Weber's diDissertationssertation focuses on the goal of building human-level artificial intelligence for real-time strategy games. It explores the research questions, What capabilities are necessary for expert RTS gameplay and how can these capabilities be integrated in a complete game playing agent? The dissertation was produced under the advisorship of Michael Mateas and Arnav Jhala in the Expressive Intelligence Studio at the University of California, Santa Cruz.

A paperback version is available on Amazon.