Evidence receipt / prediction
Published · transcript-backedTim Scarfe: prediction
28 Aug 2025 Machine Learning Street Talk Michael Timothy Bennett: Defining Intelligence and AGI Approaches
“Yeah. And and if I understand correctly, I I think that I've spent years since I read that 2007 paper, but but it it it was about an agent minimizing common complexity, which can, like, do well on a on a the the expected performance on a wide range of environments.”
Source trail
Everything needed to verify it.
- Speaker
- Tim Scarfe
- Attribution
- Verified speaker
- Claim type
- prediction
- Recorded
- 28 Aug 2025
- Publisher
- Machine Learning Street Talk
Transcript context
…Oh, yeah. And then the generalization difficulty is is like the relative entropy from that scope of tasks for, you know, from the the wider scope of tasks. Yeah. And and if I understand correctly, I I think that I've spent years since I read that 2007 paper, but but it it it was about an agent minimizing common complexity, which can, like, do well on a on a the the expected performance on a wide range of environments. The definition of task here is important. I I agree with Chelette that the tasks are what's important. I just disagree about what constitutes a task. So something like Legg and Huddle's definition with the environments and the goals and the sort of and it's the reinforcement learning framework. You've got things like actions. These are all high level abstractions we humans use to simplify the world. And as the problem of relative complexity in an interactive setting kind of illustrates, if I use a different set of abstractions to achieve the same ends, I can make it I can make something very difficult or very easy. So if I'm trying to talk about tasks, then I need to talk about embodiment as well. I can't just rely on this idea of a software mind. Because whatever the software mind does depends on the interpreter or hardware. You have to look at the system as a whole. And in cognitive science, they've got this idea of an active cognition, which is not just embodied, but in the environment as well. Because if you change the environment, then the same goals change difficulty. Which is kind of an idea that you can see in Legg and Hutter's definition. But it is something that needs to be formalized as part of the process of intelligence. Because if you just sort of assume you have a set of actions, or assume an environment, you're kind of bypassing a lot of what what intelligence needs to do to solve a task. And so…
Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.