A Worked Example of the Bayesian Mechanics of Classical Objects

Hello, everyone. Welcome. This is ACT-INF livestream number 49.0. We're discussing the paper, A Worked Example of the Bayesian Mechanics of Classical Objects. Welcome to the Active Inference Institute. We're a participatory online institute that is communicating, learning, and practicing applied active inference. This is a recorded and an archived livestream, so please provide us with feedback so we can improve our work. All backgrounds and perspectives are welcome and we'll be following video etiquette for livestreams. Head over to activeinference.org to learn more about participating in different projects and learning groups at the institute. Well, we're here today in livestream number 49.0 where we're providing some background and context for the paper, A Worked Example of the Bayesian Mechanics of Classical Objects. This paper was submitted to the Archive Preprint Server in June 2022, and we're going to be discussing and looking at a slightly revised version that was uploaded earlier this month on the 6th of September. It's a single author paper by Dalton A. R. Saktivadevel. And this video, although all .zeros are, this one is too. It's an introduction and just a preliminary discussion of some of the ideas. It's not a review or synthesis or a final word. And we'll explain a little bit more about our .zero approach in a couple of slides. But suffice to say that with Jakob, Ali, and also Ander, we had a great time over the last two months really pulling back and trying to think deeply about how to frame this paper and really this whole area of research and development because there's a lot happening. There's a lot being tied together. And so we hope that some of that sense making translates into the ways that we've laid out 49.0. And we're very much looking forward to having Dalton join for the .1 and the .2. So Dalton, thanks for all the work and for listening to this .0. Let's head to introductions and warmups. So we can just say hello and any other features of what we're excited about in this discussion. I'm Daniel. I'm a researcher in California. And I think there's a lot that is exciting about this paper. I didn't have much of a physics background before heading towards active inference in the free energy principle. So taking a first look at a lot of these physics ideas was interesting, like seeing them as part of something that already could or had been integrated rather than starting with the puzzle pieces, I think was a little bit interesting. And I'll pass to Ali. Hi, I'm Ali. I'm an independent researcher from Iran. Actually, what gravitated me towards active inference paper, this, I mean, FEP or active inference literature, and particularly this paper was, you see, I was always, I've been fascinated by doing the kind of elegant abstractions on some physical concepts or some doing doing a kind of bird's eye view over the methodological way of doing physical modeling. And I believe this paper and the other related papers offered by Dalton Sack de Vadevelle and the other and his other colleagues from Bursa's lab are the epitome of this kind of FEP abstraction in terms of FEP abstraction in terms of elegantly formalizing all the rigorous mathematical foundations of it. And yeah, that's why I'm pretty excited to be here. Actually, it's been a very fascinating and thrilling journey during the past couple of months since we began discussing this paper. And so I'm very much looking forward to it. Hello, I'm Jacob. I'm a student and researcher in the UK. There are also many people who are not able to do this. Okay, lost you up for a second. We'll just wait. We'll take a breath. Okay. We'll continue. Sorry. I got up. Go for it. Just start again. It's all good. Where was I cut off? Just start at the beginning. Okay. I got up. I got up. I got up. I got up. I got up. I got up. I got up. I got up. I got up. I got up. I got up. I got up. I got up. Okay, sorry, I got randomly disconnected. So hello, I'm Jakub, I'm a student and…