Lecture series on Causal Sets approach to QG

[marcus]

I heard that Fay Dowker (Imperial College London) is scheduled to give a series of introductory lectures on CS-quantum gravity at Perimeter in October.

They've been posting video of talks like this on PIRSA fairly consistently. So I expect Dowker's talks will be online--a short introductory video textbook.

Joe Henson just gave a one hour introduction to Dowker's introduction Henson is a postdoc at Perimeter who is an expert in the CS approach and also works in CDT (causal dynamical triangulations). One thing he has done recently is develop supercomputer or cluster tools for numerical QG work---including CDT simulations but also something called "Cactus CausalSets". Cactus is a system for distributed computing.

I thought Henson's brief one hour talk on the CS essentials was good. A place to start if you want to follow Dowker's talks later---or a clear summary of CS if that's all you want. The video is on PIRSA

http://pirsa.org/10090092
An invitation to an invitation to causal sets
Joe Henson
14 September 2010
"A brief review of some recent work on the causal set approach to quantum gravity. Causal sets are a discretisation of spacetime that allow the symmetries of GR to be preserved in the continuum approximation. One proposed application of causal sets is to use them as the histories in a quantum sum-over-histories, i.e. to construct a quantum theory of spacetime. It is expected by many that quantum gravity will introduce some kind of fuzziness uncertainty and perhaps discreteness into spacetime, and generic effects of this fuzziness are currently being sought. Applied as a model of discrete spacetime, causal sets can be used to construct simple phenomenological models which allow us to understand some of the consequences of this general expectation."

 

[marcus]

It's possible that fast massive parallel computing changes the QG picture somewhat---making some discretized approaches like Causal Sets and CDT more viable---even though they are difficult to treat analytically (by sets of equations you can solve by hand.)

Not radically--symbolic equations are still the main handle on the world--just a slight marginal change. So we might be going to hear a bit more about Causal Sets, as for instance here:

A PhD student of Fay Dowker (London Imperial) posts her dissertation:

http://arxiv.org/abs/1009.1593
Causal Set Phenomenology
Lydia Philpott
PhD Thesis, 145 pages
(Submitted on 8 Sep 2010)
"Central to the development of any new theory is the investigation of the observable consequences of the theory. In the search for quantum gravity, research in phenomenology has been dominated by models violating Lorentz invariance (LI) -- despite there being, at present, no evidence that LI is violated. Causal set theory is a LI candidate theory of QG that seeks not to quantise gravity as such, but rather to develop a new understanding of the universe from which both GR and QM could arise separately. The key hypothesis is that spacetime is a discrete partial order: a set of events where the partial ordering is the physical causal ordering between the events. This thesis investigates Lorentz invariant QG phenomenology motivated by the causal set approach. Massive particles propagating in a discrete spacetime will experience diffusion in both position and momentum in proper time. This thesis considers this idea in more depth, providing a rigorous derivation of the diffusion equation in terms of observable cosmic time. The diffusion behaviour does not depend on any particular underlying particle model. Simulations of three different models are conducted, revealing behaviour that matches the diffusion equation despite limitations on the size of causal set simulated. The effect of spacetime discreteness on the behaviour of massless particles is also investigated. Diffusion equations in both affine time and cosmic time are derived, and it is found that massless particles undergo diffusion and drift in energy. Constraints are placed on the magnitudes of the drift and diffusion parameters by considering the blackbody nature of the CMB. Spacetime discreteness also has a potentially observable effect on photon polarisation. For linearly polarised photons, underlying discreteness is found to cause a rotation in polarisation angle and a suppression in overall polarisation."

In the acknowledgments, the author thanks Joe Henson for making "Cactus CausalSets" simulation available. It seems to be how she got a good many of her results.

Note her statement that Causal Sets is a Lorentz invariant candidate theory of QG.

 

[marcus]

The promised minicourse, or lecture series, on Causal Sets is scheduled to take place in about a week---18-22 October
http://www.perimeterinstitute.ca/Scientific/Courses/An_Invitation_to_Causal_Sets/

It will be taught by both Rafael Sorkin and Fay Dowker.

They are the senior experts in Causets. Should be a definitive treatment.

 

[atyy]

This is one of the most beautiful approaches to QG - the other one is Bilson-Thompson's - though I think both will never work (it's going to be string theory, unfortunately;)

Edit: My favourite approaches are still Markopoulou and Wen, but somehow the adjective for those doesn't seem to be "beautiful".

 

[marcus]

This is one of the most beautiful approaches to QG - the other one is Bilson-Thompson's - though I think both will never work (it's going to be string theory, unfortunately;)

Yes, I see you are always standing up for string theory. I don't pick winners (don't think we can know the future) instead I try to follow the action and understand the reasons for current progress (as best I can.)

You often help me by pointing out details I haven't noticed---papers I've missed. Like the AQG IV Thiemann Giesel--which was very interesting!

I would say that Causets has not made much progress so far because it is "too" discrete so that one cannot calculate anything analytically (with equations).
However the appearance of massive simulation by computers has changed the rules somewhat. Causets might be poised to make some noticeable advances.

I would say that B-T requires a manifold for things to be embedded in, so they can braid and knot. The program stalled in 2008. I await further news.

You have reminded me that Thiemann ditched the manifold already in 2007.

And he embraced the Cosmic Microwave Background---or the "dust" if you prefer---or the Friedmann model coordinates of the universe.

That, I confess, appeals to me temperamentally, but I would not dare say it is the right thing to do.

What HAS seemed the right thing to do is to ditch the manifold, which Thiemann did in 2007 and which Rovelli's team seems to be rather thoroughly doing in 2010. I don't say that it is ultimately the correct move, but I'm seeing that it is accompanied by an enormous amount of progress. 2010 is a banner year for LQG.

Also Rovelli has an inconspicuous way to address the time problem which does not depend on dust but which gives back the Friedmann/CMB idea of time. That means there is a kind of effective compatibility, which might work everywhere except on the most fundamental level. Just speculating.

Anyway, this thread is about Causets, and the signal thing about Causets is that it was the first completely manifoldless approach.

Sorkin appears brilliant because on philosophical grounds he did in 1990 what everybody else is doing around 2010, twenty years later. He decided that spacetime must be realized entirely without a differential manifold, or even a topological continuum, because those are merely aids to doing calculus. In Sorkin's philosophy the essentials are, as Einstein hinted, the events and they are related causally.

Rovelli's philosophy is somewhat different and leads to spin networks instead of a partial order discrete set of events. I think in that philosophical view it is measurement that is fundamental. A spin network is essentially a scheme of geometric observation and measurement (areas, volumes, etc.) that leads us to the idea of a "state" of geometry.
It took that idea longer to hatch out of the manifold, it was more "embryonic" so to speak and also more complex, so has taken longer to gestate.

We don't know that any of these ideas are right, we can only judge rates of gainful innovation.

 

[Carey]

Check out the "Discussion" tab for the Wikipedia entry on causal sets. There are relative frequency ratios inherent in causal sets, and these can serve as energy ratios in accord with Planck's E=hf. Thus causal set theory offers a simple explanation of the origin of mass, in terms of mass-ratio. It is then a simple matter to construct space-time and its particle-like sequences as standalone causal sets. -- Carey R. Carlson

 

[marcus]

Carey, thanks for renewing interest in Causal Sets and reviving this thread!

The lecture series that we heard was planned (and prompted this thread) did actually occur and is here:
http://pirsa.org/C10020

The series of 5 lectures is called *An Invitation to Causal Sets* and it has portions by both Rafael Sorkin and Fay Dowker

 

[Fra]

It seems this thread may contain some answers I posed in the other thread. I'll skime those links later and see if they may serve as the review I was looking for.

I'm intrigued by the common denominators of several programs.

/Fredrik

 

[Fra]

My memory must be short. I started watching that video and soon recognized that I did watch it already some months ago. I also remember that I the way it was presented it wasn't a fundamental enough rethinking as the main point was to simplify say the PI calculation, while I think the entire computational algorithm of the PI (including the quantum logic) should follow from the reconstruction.

I guess I think looking for something more radical, that explains also how the the 4D structure emerges as evolved codes from a simply one-dimensional order, and why the expectation algorithm looks like the feynmann PI. Somehow my gut feeling is that there is a deeper explanation of this.

/Fredrik