2

New to LessWrong?

New Comment
37 comments, sorted by Click to highlight new comments since: Today at 3:24 PM

about the problems with wikipedia

The problem that Wikipedia adopts standards from modern evidence-based medicine? It's better to read a meta-analysis from Cochrane (which is a secondary source) than reading various papers that make statements about what a drug did that might not replicate.

Edit: Down the rabbit hole...

An interview with the founder, pretty interesting and straightforward

https://www.quantamagazine.org/20150604-quantum-bayesianism-qbism/

I am leaning to the hologram outlook lately. Still enjoy Rovelli's writing more than most tho...

Picked this up in a thread at Artic Sea Ice site, that i haunt for methane related info, but the Holo Dark Energy/Entropy model is a fabulous cosmological model.

Michael Paul Gough (2013), "Holographic Dark Information Energy: Predicted Dark Energy Measurement", Entropy, 15(3), 1135-1151; doi:10.3390/e15031135

http://www.mdpi.com/1099-4300/15/3/1135?trendmd-shared=0&utm_campaign=trendmd&utm_medium=cpc&utm_source=trendmdwidget

Abstract: "Several models have been proposed to explain the dark energy that is causing universe expansion to accelerate. Here the acceleration predicted by the Holographic Dark Information Energy (HDIE) model is compared to the acceleration that would be produced by a cosmological constant.

and:

Michael Paul Gough (2014), "A Dynamic Dark Information Energy Consistent with Planck Data", Entropy, 16(4), 1902-1916; doi:10.3390/e16041902

http://www.mdpi.com/1099-4300/16/4/1902?trendmd-shared=0&utm_campaign=trendmd&utm_medium=cpc&utm_source=trendmdwidget

Abstract: "The 2013 cosmology results from the European Space Agency Planck spacecraft provide new limits to the dark energy equation of state parameter. Here we show that Holographic Dark Information Energy (HDIE), a dynamic dark energy model, achieves an optimal fit to the published datasets where Planck data is combined with other astrophysical measurements. HDIE uses Landauer’s principle to account for dark energy by the energy equivalent of information, or entropy, of stellar heated gas and dust. Combining Landauer’s principle with the Holographic principle yields an equation of state parameter determined solely by star formation history, effectively solving the “cosmic coincidence problem”.

source post is fun too http://forum.arctic-sea-ice.net/index.php?topic=1578.msg91730#new

Arxiv papers for Fuch's from comments in interview article

QBism and the Greeks: why a quantum state does not represent an element of physical reality

https://arxiv.org/abs/1412.4211

QBism, the Perimeter of Quantum Bayesianism

https://arxiv.org/abs/1003.5209

QBism is very interesting, but still lacks a lot of foundations for it to be taken seriously as a mainstream interpretation: why SIC-POVMs should be preferred? What they can tell us about the ontology of quantum mechanics? Do they exists in all dimensions? How are they related to the Heisenberg picture?

Q: Quantum. Bayesianism isn't the LessWrong official preferred interpretation of QM because....?

In the Sequence, Eliezer made a strong case for the realist interpretation of QM (neo-Everettian many worlds), based on decoherence and Occam's razor. He then, in another point of the Sequence, tied that problem with interesting questions about anthropic probability (the infamous anthropic trilemma), and that cemented MWI as the preferred way to think about QM here.

On the other hand, I think we are still missing the big picture about quantum mechanics: ER = EPR, categorical quantum mechanics, QBism etc. all points us to interesting unexplored directions.

In the Sequence, Eliezer made a strong case for the realist interpretation of QM (neo-Everettian many worlds), based on decoherence and Occam's razor.

It's tendentious to call MWI the only realistic interpretation.

EY makes a case against CI, which in most circumstances would be a case against anti-realism. However his version of CI is actually OR, another realistic theory. So he never makes a case for realism against irrealism.

As far as I know, neoEverett is the smallest realist interpretation: Eliezer argued not only against anti-realism, but also in favor of the smallest theory that falls out of the formalism.

But MWi looks huge compared to RQM: it reifies basis, which is much more naturally explained as a choice by an observer, ie a "map" feature.

There are a number of kinds and grades of non-realism. Objective collapse theories reify both state and collapse, MWI refies state only and RQM refies neither. Nonethless, it is not a completely anti-realist theory.

As far as I know, RQM is not even a complete interpretation of quantum mechanics. In the original paper by Rovelli, there are many holes left which I thought nobody has patched yet. If you know of an exposition that corrects those problems, I would gladly read it.

An incomplete interpretation that is in the right lines may be better than a complete one that is not.

You have to value elegance more than correctedness, though.
I'm not say that RQM is incorrect, but I am saying that until it's completed, nobody can tell if it's correct.
Also, nobody can guarantee that when completed it won't carry more weight than neoEverett.

I'm not putting g a 100%..sorry, 99,99999% weighting on RQM. But its very existence undermines EYs argument for MWI,because it suggests third alternatives to a number of alleged either/or dichotomies

Do you think that we're likely to find something in those directions that would give a reason to prefer some other interpretation than MWI?

My idea is more on the line of "in the future we are going to grasp a conceptual frame that would make sense of all interpretations" (or explain them away) rather than pointing to a specific interpretation.

If it doesn't fundamentally change quantum mechanics as a theory, is the picture likely to turn out fundamentally different from MWI? Roger Penrose, a vocal MWI critic, seems to wholeheartedly agree that QM implies MWI; it's just that he thinks that this means the theory is wrong. David Deutsch, I believe, has said that he's not certain that quantum mechanics is correct; but any modification of the theory, according to him, is unlikely to do away with the parallel universes.

QBism, too, seems to me to essentially accept the MWI picture as the underlying ontology, but then says that we should only care about the worlds that we actually observe (Sean Carroll has presented criticism similar to this, and mentioned that it sounds more like therapy to him), although it could be that I've misunderstood something.

If it doesn't fundamentally change quantum mechanics as a theory, is the picture likely to turn out fundamentally different from MWI?

CI/OR is a different picture to MWI, yet neither change QM as a number-crunching theory. You have hit on the fundamental problems of empiricism: the correct interpretation of a data is underdetermined by data, and interpretations can differ radically with small changes in data or no changes in data.

I'm not sure what you mean by OR, but if it refers to Penrose's interpretation (my guess, because it sounds like Orch-OR), then I believe that it indeed changes QM as a theory.

These are difficult question because we are speculating about future mathematics / physics.

First of all, there's the question of how much of the quantum framework will survive the unification with gravity. Up until now, all theories that worked inside it have failed; worse, they have introduced black-hole paradoxes: most notably, thunderbolts and the firewall problem. I'm totally in the dark if a future unification will require a modification of the fundamental mathematical structure of QM. Say, if ER = EPR, and entanglement can be explained with a modified geometry of space-time, does it mean that superposition is also a geometrical phoenomenon that doesn't require multiple worlds? I don't really know.

But more on the point, I think (hope?) that future explorations of the quantum framework will yield an expanded landscape, where interpretations will be seen as the surface phoenomenon of something deeper: for example, something akin to what happens in classical mechanics with the Hamiltonian / Lagrangian formulations.

On a side note, I've read only the Wikipedia article on QBism and my impression was that it had an epistemological leaning, not ontological: if you use only SIC-POVMs, you can explain all quantum quirks with the epistemology of probability distributions. I might be very wrong, though.

Fair enough. I feel like I have a fairly good intuitive understanding of quantum mechanics, but it's still almost entirely intuitive, and so is probably entirely inadequate beyond this point. But I've read speculations like this, and it sounds like things can get interesting: it's just that it's unclear to me how seriously we should take them at this stage, and also some of them take MWI as a starting point, too.

Regarding QBism, my idea of it is mostly based on a very short presentation of it by Rüdiger Schack at a panel, and the thing that confuses me is that if quantum mechanics is entirely about probability, then what do those probabilities tell us about?

it's just that it's unclear to me how seriously we should take them at this stage

Well, categorical quantum mechanics is a program under developement since 2008, and it gives you a quantum framework in any computational theory with enough symmetries (databases, linguistics, etc).
It spawned quantum programming languages and a graphical calculus. So I think it's pretty succesful and has to be taken seriously, albeit it's far from being complete (it lacks a unified treatment of infinite systems, for example).

Depends what you mean by "about". The (strong) Qbist perspective is that probabilities, including those derived from quantum theory, represent an agents beliefs concerning his future interactions with the world. If you're looking for what these probabilities tell us about the underlying "reality" then that's an open question, which Fuchs et al are still exploring.

If you're looking for what these probabilities tell us about the underlying "reality"

I am. It seems to me that if quantum mechanics is about probabilities, then those probabilities have to be about something: essentially, this seems to suggest that either the underlying reality is unknown, indicating that quantum mechanics needs to be modified somehow, or that Qbism is more like an "interpretation of MWI", where one chooses to only care about the one world she finds herself in.

The QBist stance is that we "know" very little about the underlying reality. One of the only things that Chris Fuchs is willing to accept as an objective property of a quantum system is its Hilbert space dimension.

I doubt it's sensible to talk about an interpretation of MWI. MWI says that the wavefunction is a real physical object and wavefunction splitting is something that's genuinely physically occurring. QBism denies that the wavefunction is a real physical object.

We've already got a number of problems with MW -- see Dowker and Kent's paper.

The question is whether there is anything better. To go back to my original question, EY appears not to have heard of QBism, RQM, and other interpretations that aren't mentioned in The Fabric of Reality.

Guess I'll have to read that paper and see how much of it I can understand. Just at a glance, it seems that in the end they propose one of the modified theories like GRW interpretation might be the right way forward. I guess that's possible, but how seriously should we take those when we have no empirical reasons to prefer them?

I guess that's possible, but how seriously should we take those when we have no empirical reasons to prefer them?

Doesn' that rebound on the argument for MWI?

Sincere and consistent instrumentalists may exist, but I think they are rare. What is much more common is for people to compartmentalise, to take and irrealist or instrumetalist stance about things that make them feel uncomfortable, while remaining cheerfully realist about other things.

At the end of the day, being able to predict phenomena isn’t that exciting. People generally do science because they want to find out about the world. And “rationaists”, internet atheists and so on generally do have ontological commitments, to the non-existence of gods and ghosts, some view about whether or not we are ina matrix and so on.

I'm certainly not an instrumentalist. But the argument that MWI supporters (and some critics, like Penrose) generally make, and which I've found persuasive, is that MWI is simply what you get if you take quantum mechanics at face value. Theories like GRW have modifications to the well-established formalism that we, as far as I know, have no empirical confirmation of.

There are modified theories, there is no unequivocal "face value".

Eliezer and E.T. Jaynes strongly urge seeing probabilities as subjective degrees of certainty that follow fixed laws (an extension of logic). If QBism is supposed to be compatible with this view - and yet not a form of MWI - then where do the complex numbers come from? Do they represent the map or the territory?

A Qbist would say they represent the map. The complex vector formalism of quantum theory is simply a convenient/elegant manual for predicting the outcomes of one's future interactions with nature. It may be able to tell us something about the territory, but is not the territory itself.

Do you see how that's not an answer? Why do they work?

A solipsist/Parmenidean theory of reality would need to add another theory, or a copy of all the evidence, in order to make predictions. It would be simpler to drop the solipsism and just state the part of the theory that does the work (which would be our account of reality). Eliezer makes the case that something like a wavefunction exists, and the splitting of 'worlds' somehow gives rise to the Born probabilities.

The QBist aim is not to provide an ontological description of the universe. Rather, it is to persuade you that whatever such a description is, quantum theory ain't it.

"The professed goal is to strip away all those elements of quantum theory that can be interpreted in subjective, agent-dependent terms. The hope is that whatever remains will hint at something essential and objective about nature."

I've read the quantum theoretic parts of the sequences: Eliezer doesn't really make a case for why Born probabilities arise. Indeed this is one of the major open problems with the MWI.

That's the basic, some say the only, mystery of MWI: why the world operates according to subjective probability?
You'll find this question posed in the Sequence in some places.

No, that is not the question I asked. The question I asked was what the god-damned imaginary numbers mean, if they aren't describing reality. Because they don't look like subjective probability.

The existence of some form of subjective probability is perfectly compatible with the existence of some other form of objective probability.

Noticed , I think its first time i have seen that, usually ask for more primary published works.....