RT @IrisVanRooij: 31. Kwisthout, J. & van Rooij, I. (2020). Computational resource demands of a predictive Bayesian brain. Computational…
RT @IrisVanRooij: 31. Kwisthout, J. & van Rooij, I. (2020). Computational resource demands of a predictive Bayesian brain. Computational…
RT @IrisVanRooij: 31. Kwisthout, J. & van Rooij, I. (2020). Computational resource demands of a predictive Bayesian brain. Computational…
31. Kwisthout, J. & van Rooij, I. (2020). Computational resource demands of a predictive Bayesian brain. Computational Brain & Behaviour, 3, 174–188. https://t.co/t6ir7JJANc
RT @bertdv0: @IrisVanRooij @SuperboffinLA hi @IrisVanRooij, your paper https://t.co/GC81qhZJJW focuses on solving NP-hard problems in polyn…
RT @bertdv0: @IrisVanRooij @SuperboffinLA hi @IrisVanRooij, your paper https://t.co/GC81qhZJJW focuses on solving NP-hard problems in polyn…
RT @IrisVanRooij: "An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ...…
RT @IrisVanRooij: "An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ...…
RT @IrisVanRooij: "An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ...…
RT @IrisVanRooij: "An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ...…
@alexandersclark @IrisVanRooij Sure, don't disagree, but see @bertdv0's points here. about why these methods are useful: https://t.co/vEaxG7zEn6
@JasonCDaniels @ChombaBupe Thanks :) There is continued discussion on the topic here, in case of interest https://t.co/EZzmaWpoDN
@IrisVanRooij Moreover, as @bertdv0 points out here, the point of using these technologies isn't really to solve NP-hard problems in polynomial time, which I agree would not be tractable. So respectfully, I remain unconvinced. I'd like to see more targeted
RT @IrisVanRooij: Hi @mjdramstead thanks for the Q. The analyses in our textbook and in this paper on predictive Bayesian brain (see https:…
Hi @mjdramstead thanks for the Q. The analyses in our textbook and in this paper on predictive Bayesian brain (see https://t.co/t6ir7JJANc) are at computational level: hence all algorithms are subject to the intractability & inapproximability results.
RT @bertdv0: @IrisVanRooij @SuperboffinLA hi @IrisVanRooij, your paper https://t.co/GC81qhZJJW focuses on solving NP-hard problems in polyn…
RT @bertdv0: @IrisVanRooij @SuperboffinLA hi @IrisVanRooij, your paper https://t.co/GC81qhZJJW focuses on solving NP-hard problems in polyn…
@IrisVanRooij @SuperboffinLA hi @IrisVanRooij, your paper https://t.co/GC81qhZJJW focuses on solving NP-hard problems in polynomial time, but the excitement about Active Inference (AIF) is not that it solves NP-hard problems in polynomial time. The promise
@SuperboffinLA We wrote a paper about it in this paper: https://t.co/t6ir7JJANc See also highlighted sentence in the abstract. https://t.co/W09Ax7xOcu
@ChombaBupe Fwiw, my take in thread https://t.co/erxWhXle9F
RT @IrisVanRooij: "An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ...…
RT @IrisVanRooij: @GordonBrianR I thought for a moment this was sarcasm & you were joking. But now that I read the open letter I see we are…
RT @IrisVanRooij: @GordonBrianR I thought for a moment this was sarcasm & you were joking. But now that I read the open letter I see we are…
RT @IrisVanRooij: @GordonBrianR I thought for a moment this was sarcasm & you were joking. But now that I read the open letter I see we are…
RT @IrisVanRooij: @GordonBrianR I thought for a moment this was sarcasm & you were joking. But now that I read the open letter I see we are…
RT @IrisVanRooij: "An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ...…
RT @IrisVanRooij: "An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ...…
RT @IrisVanRooij: "An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ...…
"An important conjecture in neuroscience is that a [Bayesian] brain can efficiently approximate Bayesian inferences. ... Yet, as we show in this paper, ... each key subcomputation invoked in predictive processing potentially hides a computationally intract
@GordonBrianR I thought for a moment this was sarcasm & you were joking. But now that I read the open letter I see we are in a rerun of history. Bayesian/active inference approaches also won't scale without crucial insights. I see no evidence of thos
@EliSennesh @PeteDaGuru @mjdramstead This work seems relevant here https://t.co/EGxQeSqhyD
@matspike @aswinpaule Given the topic, I'd recommend a look at this paper https://t.co/EGxQeSqhyD
RT @EdSacredProfane: The reality is memory is not the residue of though but the residue of thought and experience plus prediction according…
The reality is memory is not the residue of though but the residue of thought and experience plus prediction according to neuroscience. https://t.co/9zfjpTdbDU
@sir_deenicus I’ll post a few that may be related 1. https://t.co/t6ir7JJANc
@neuro_data .. also a comp-level theory can be decomposed into multiple sub comp-levels theories. For examples of this see e.g. this paper with @JohanKwisthout https://t.co/t6ir7JJANc
RT @NoahGuzman14: @rhamdu That’s a good question. I don’t exactly know how you would go about comparing efficiency though? These works by @…
@rhamdu That’s a good question. I don’t exactly know how you would go about comparing efficiency though? These works by @IrisVanRooij might be a good place to start: Complexity of predictive processing: https://t.co/tZ40UhYQZ7 Complexity of self-organiz
@the_aiju it implodes and explodes :) https://t.co/t6ir7JJANc
RT @IrisVanRooij: @vascobrazao @samuelmehr @aeryn_thrace @o_guest Fwiw, we gave this feedback already last year, when the paper was out as…
RT @IrisVanRooij: @vascobrazao @samuelmehr @aeryn_thrace @o_guest Fwiw, we gave this feedback already last year, when the paper was out as…
@vascobrazao @samuelmehr @aeryn_thrace @o_guest Fwiw, we gave this feedback already last year, when the paper was out as preprint. Not sure why it was ignored? (see this tweet and whole thread https://t.co/gqdcEIj97E)
If the bioinformatics case is too unfamiliar for you, see a cogsci perspective here. https://t.co/gcf2aYaSFb
@VandekerckhoveJ @zerdeve @modrak_m @BrianNosek But what if I have nothing to preregister? What if my method is mathematical proof? Then I cannot publish in the "best" journals? Example: https://t.co/t6ir7JJANc
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
@DaltonSakthi @computa_shrink @JohanKwisthout I think you may be referring to this work by @JohanKwisthout and myself? Personally, I do not see this as an argument against PP, but more as a theoretical challenge https://t.co/t6ir7JJANc
RT @NeuroYogacara: Been banging my head against some stuff, then I remembered this paper by @JohanKwisthout and @IrisVanRooij, which I will…
RT @DBicgroup: Happy to announce that the paper "Computational Resource Demands of a Predictive Bayesian Brain" by @IrisVanRooij and @Johan…
(Predictive processing can be potentially very resource demanding, however: https://t.co/EGxQeSqhyD. Cost-effectiveness can thus not a be assumed without explanatory challenges) 9/
@VandekerckhoveJ @JCSkewesDK @JonMcphetres @zerdeve @djnavarro Even the field-driving kind may not (consistently) be referred to as "theory". Did a quick check on one of my own papers: https://t.co/t6ir7JJANc While "Predictive Processing" is very a influen
RT @IrisVanRooij: @qarthage This paper discusses the problem in a cognitive neuroscience context, but many of the complexity-theoretic resu…
@qarthage This paper discusses the problem in a cognitive neuroscience context, but many of the complexity-theoretic results build on theorems in computer science https://t.co/t6ir7JJANc
@NoahGuzman14 @lastpositivist I see it is from 2017, so will not include more recent mathematical results, like these: https://t.co/EGxQeSqhyD
RT @IrisVanRooij: 🧵about: Kwisthout, J. & van Rooij, I. (2019). Computational resource demands of a predictive Bayesian brain. Computatio…
RT @DBicgroup: Happy to announce that the paper "Computational Resource Demands of a Predictive Bayesian Brain" by @IrisVanRooij and @Johan…
RT @IrisVanRooij: 🧵about: Kwisthout, J. & van Rooij, I. (2019). Computational resource demands of a predictive Bayesian brain. Computatio…
🧵about: Kwisthout, J. & van Rooij, I. (2019). Computational resource demands of a predictive Bayesian brain. Computational Brain & Behaviour. https://t.co/NFkimVGX1P https://t.co/EGxQeSqhyD
RT @DBicgroup: Happy to announce that the paper "Computational Resource Demands of a Predictive Bayesian Brain" by @IrisVanRooij and @Johan…
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @IrisVanRooij: Want to get a quick view of the motivations and highlights of the paper, check out this thread by @JohanKwisthout: https:…
RT @IrisVanRooij: Want to get a quick view of the motivations and highlights of the paper, check out this thread by @JohanKwisthout: https:…
@LizBonawitz You are one of few 🙂 I have more thoughts to share but that takes more time, because there are conceptual issues to untangle. Before I sign off for the night, I thought I'd note that we briefly also cover approximation by sampling in @CompBra
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @JhendersonIMB: Important paper for those interested in the predictive processing account of the brain. https://t.co/hX36nTbvEp
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @CompBrainBeh: now out in @CompBrainBeh @JohanKwisthout and @IrisVanRooij argue that the jury is still out about whether the cognitive c…
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
V interesting
RT @IrisVanRooij: Wonderful to see the interest in this work! One thing I’d like to clarify though as I’m seeing some misunderstandings:…
RT @JhendersonIMB: Important paper for those interested in the predictive processing account of the brain. https://t.co/hX36nTbvEp
RT @patchurchland: This is an important result challenging the fashionable predictive coding story: https://t.co/pEuesHxyAf
@henk_aart Looks like I'll be reading that paper! https://t.co/rEeAiWEJPv Looks quite nice with diagrams much like mine on the subject, and little maths. Bizarre though that they decompose prediction into several computationally intractable problems🤔