Session details
Date: Apr 22, 2024
Series: GuestStream #080.1
Guests: Laura Desirée Di Paolo
Paper: Active Inference Goes to School. The Importance of Active Learning in the Age of Large Language Models
यह पृष्ठ मशीन द्वारा अंग्रेज़ी से हिंदी में अनुवादित किया गया था। अंग्रेज़ी मूल देखें
GuestStream #080.1
Apr 22, 2024 · with Laura Desirée Di Paolo
▶ Watch on YouTube ↗Date: Apr 22, 2024
Series: GuestStream #080.1
Guests: Laura Desirée Di Paolo
Paper: Active Inference Goes to School. The Importance of Active Learning in the Age of Large Language Models
Transcript
The full transcript is available on GitHub. This excerpt is generated by automated speech recognition and may contain errors.
all right hello it is April 22nd 2024 we're in active inference guest stream 80.1 with Laura desire Deo talking about active inference Goes to School the importance of active learning in the age of llms we'll have a presentation than some discussion so thank you Laura for joining looking forward to this thank you very much for having invited me um okay so I'm going to present practically one of our latest publication which is um in fact that the inference Goes to School uh oh wait I have to Yes um this is an overview of the talk more or less I'll try to go as fast as I can for the first part because I'd like to spend a little bit more time on the later slides and I'm sorry for some misspelling or mistakes that you will find here and there in the in the images because I had long fights with Jud for trying to uh um generate images that would vate better my my my talk and at some point I after 50 or 60 uh attempts I simply gave gave up and they choose just the best one that could I could uh I could achieve to I could achieve okay so ah still I have to use my mouse uh just few things about myself I've been uh interested in learning since I uh uh I started University maybe because I was uh considered a slow learner in uh in high school definitely not an high achiever uh um and therefore when I started to uh to study at University I started to study learning different kind of learning different learning across species and the evolution of learning and so on and so forth uh till the point in which I started to be interested in learning in educational settings thinking okay this is the perfect recipe for Success because you have all the ingredients that you might have you have super curious beings for whosoever had to deal with children before they uh go to school uh knows that they um have this insane curiosity and they ask wise about everything theoretically in school you should have tons of learning opportunities and you finally have people that are trained to give uh the answers that the children seek to have unfortunately the sit the situation in school is not always so nice for quoting the um principal a British principal and during an interview who said it's incredible that children prefer to be home sick that going to school the place in which they can learn stuff consider that learning is fun and for what concerns me I am a researcher and as I assume most of us um I love learning I think learning very very fun so I feel part of my my job or my duty to bring back the fun into learning particularly inal educational settings or in compulsory education um this is just a slide with some uh basic literature about active inference and and learning I won't be too specific about active inference uh there is here a that eventually could um could follow up for some questions more specific about the active inference because it's way more prepared than I am but anyway this is like um I don't know if you have I have to move this one okay perfect um uh this is a basic model of active uh inference the idea of cognition in active inference for which the agent and the cognitive system of the agent is not waiting for the information to arrive to fill in but instead uh act proactivity so makes predictions about the state of the word in which expect to leave the information that that expect to uh to perceive into the word and then every time compare the predictions that has made with the effective information the effective situation of the world in that particular moment and when there is a discrepancy discrepancy uh in in terms of both surprise or events that are unexpected or errors uh either update the model for making better predictions or change the words and so act onto the word for uh the orinal predictions in this model of cognition learning is a sort of imperative because every time that there is a discrepancy between the prediction and the state of the word is like a red flag that says okay here there is something that can be learned…