In this episode, Gianluca and Jared discuss the neurobiology of predictive processing, the role of dopamine in reward signalling, satisfaction, and human desire. They also touch on the relation of reward to reinforcement learning and artificial intelligence. The podcast then shifts to discuss the clinical manifestations of the predictive brain, including the computational psychiatry of depression, autism-spectrum-disorder, and schizophrenia, and how top-down prediction, and precision-weighted prediction error explain many important features of these conditions.
This week’s episode of Bit of a Tangent is a continuation of our 3 part series on Predictive Processing - a fascinating neuroscientific account of the way our brains come to understand the world. If you haven’t already listened to part 1 (episode 28), check that out first! In this episode Gianluca and Jared discuss the importance of sensory precision, the circular causality of prediction and action, how action comes about in a counterintuitive way from high confidence predictions, the way in which prior expectations can bias sensory sampling, and how this could lead to self-fulfilling spirals that would make your life either much worse or much better! Find us on Twitter @podtangent
It's here. It's finally here! The first in a series of three episodes diving into the predictive processing theory of brain function! Jared and Gianluca discuss top-down versus bottom-up perception, Bayesian Brain hypotheses, the logic of caring only about prediction error, and the phenomenology of visual illusions!
Gianluca and Jared have survived 2020 (so far) and are back for Season 3 of Bit of a Tangent. In this episode they bring you 7 new habits and techniques that can be used to iteratively upgrade yourself — even in lockdown. Forget everything else that’s going on in the world, and take a deep dive into personal optimisation. Or, as they’d put it, prepare to geek out on organisation hacks, bootstrapped learning, and motivation pumps. Shownotes at podtangent.com
[Epistemic status: slightly intoxicated] Jared and Gianluca exchange rapid-fire questions whilst becoming increasingly inebriated. Think of this episode as the Ferriss/Rose “Random Show,” but with more references to Bayesian inference. It’s pretty likely to entertain you, but hopefully you’ll find it interesting too. Topics include Email response times, VR projects, and the statistical approach to an ethical diet. -------- Shownotes: -------- Jared on Twitter: www.twitter.com/jnearestn Gianluca on Twitter: www.twitter.com/QVagabond Bit of a Tangent on Twitter (www.twitter.com/podtangent) and Instagram (instagram.com/podtangent/) Last episode: https://www.podtangent.com/e/025-self-supervised-machine-learning-introduction-intuitions-and-use-cases/ The Game Changers: www.imdb.com/title/tt7455754/?ref_=fn_al_tt_1 Joe Rogan Experience #1389 - Chris Kresser Debunks "The Gamechangers" Documentary: www.youtube.com/watch?v=Dq4Apc2Xk7Q Sam Altman on Conversations with Tyler: conversations...
On this episode of Bit of A Tangent, we discuss the emerging field of self-supervised machine learning. This is an immensely exciting area of active research in machine learning and AI - one which most people haven’t even heard about yet! We build up to the intuition for the topic by covering supervised and unsupervised learning; autoencoders and dimensionality reduction, and exploring how these techniques could be applied to Gianluca’s Quantified Self n=1 sleep quality dataset. We culminate in a detailed discussion of the state-of-the-art Contrastive Predictive Coding model, and how it allows us to learn about the structure of the world, without tonnes of labelled training data! -------- Shownotes: -------- Jared on Twitter:www.twitter.com/jnearestn Gianluca on Twitter:www.twitter.com/QVagabond Bit of a Tangent on Twitter (www.twitter.com/podtangent) and Instagram (instagram.com/podtangent/) Summer school on Computational Neuroscience: http://imbizo.africa/ Control problem in AI:...
Gianluca announces his latest project—the audio version of Nate Soares’ Replacing Guilt series—and Jared explains why it’s a notable contribution to the rationalist community. Together, they discuss how to gain the most from listening to the series and why it’s something you should strongly consider doing if you often find yourself binging Netflix until 3 AM.Shownotes at podtangent.com
Jared and Gianluca have an uncompromising exchange about computational neuroscience and its implications for life and AI research. Shownotes at podtangent.com
Gianluca presents his much-requested three non-fiction and three fiction recommendations for 2020. Jared recommends a book so long and gripping that you should probably avoid it if you have anything productive to do in the next six months.Shownotes at podtangent.com
Bit of a Tangent is back! In this episode, Jared shares the 19 ideas he couldn’t stop thinking about in 2019, whilst Gianluca does two B-grade Joe Rogan impressions. Together, they discuss everything from mechanism design to personal fitness hacks, and throw in some recommendations for supercharging your learning in the new year. Strap yourselves in for a taste of the fascinating conversations that lie ahead in 2020! Find shownotes at podtangent.com