Hi there, I'm Cam Allen
I'm a postdoc studying AI with Stuart Russell at the Center for Human-Compatible Artificial Intelligence (CHAI) at UC Berkeley. In spring 2024, I was also a lecturer for CS 188, Berkeley's upper-division artificial intelligence course. Previously, I completed my PhD at Brown University, advised by George Konidaris.
My research goal is to steer artificial intelligence towards human flourishing. The challenge is that “human flourishing” is meaningless outside of a reference frame, and there are many valid reference frames to choose from.
In my work, I develop frame-aware AI systems that not only operate within a given frame, but also recognize the frame, make it explicit, propose alternatives, navigate between frames as context demands, and help humans do the same. I envision a future where people and AI can collaborate to notice and transcend the assumptions hidden in their reference frames and then use these insights to improve themselves and the world around them.
I am interested in reframing the foundations of intelligence—the computations that enable agency, learning, planning, abstraction, interaction, and introspection—to account for reference frames explicitly:
- Definition. How can agents construct appropriate conceptual models of their decision problem? How should agents form representations of what they know?
- Inspection. What representations and conceptual models do AI systems already learn implicitly, and can we identify similar mental structures in humans?
- Transformation. How can AI systems define new reference frames on top of existing ones? Abstraction and attention can narrow the frame by ignoring irrelevant information. What would it look like to expand the frame?
- Selection. After generating alternative problem framings, how should an AI system decide which is appropriate?
- Coordination. How might we align the reference frames of different decision-making agents? How can we learn human-interpretable state representations—and skills that precisely manipulate them? How can we account for unknown or changing reference frames?
- Introspection. What makes an agent self-aware? How can it notice its own reference frame? Is there a link between self-awareness and experience? Experience and meaning? How can we build agents that help us have more meaningful and beautiful experiences?