What is the true purpose of inflation in the early universe?

[Torbjorn_L]

Correct me if I'm wrong, but these quantitative aspects of the theory do not just fall out from our existing confirmed theories and known fields.

Agreed.

That an inflation theory, thus formulated, can explain many things does not make it correct (does it?).

Agreed. What makes it correct is if the contenders die: "And as of now, we do not have any well-developed explanations that compete with inflation, despite many attempts."

How is it possible to claim this when the inflation parameters appear to be chosen to make these "predictions"?

How else would it be done? In measurement theory you do hypothesis testing on observations (and so hypotheses and theories). The observation and the constraints (experimental and chosen) are tested in combination.

If there are free parameters, they (as the rest) are subject to verification. E..g. WMAP, Planck and many more observatories several data releases are consistent with each other and previous observations.

Another test for robustness is that LCDM cosmology is the first self-consistent cosmology. Rip inflation out, and that falls apart.

In the words of the eminent empiricist SH: "How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?" [ http://en.wikiquote.org/wiki/Sherlock_Holmes ] LHC proved that Conan Doyle was correct:

"The Laws Underlying The Physics of Everyday Life Are Completely Understood

Not sure why people don’t make a bigger deal out of this fact. ...

A hundred years ago it would have been easy to ask a basic question to which physics couldn’t provide a satisfying answer. “What keeps this table from collapsing?” “Why are there different elements?” “What kind of signal travels from the brain to your muscles?” But now we understand all that stuff. (Again, not the detailed way in which everything plays out, but the underlying principles.) Fifty years ago we more or less had it figured out, depending on how picky you want to be about the nuclear forces. But there’s no question that the human goal of figuring out the basic rules by which the easily observable world works was one that was achieved once and for all in the twentieth century.

You might question the “once and for all” part of that formulation, but it’s solid. Of course revolutions can always happen, but there’s every reason to believe that our current understanding is complete within the everyday realm."

[ http://blogs.discovermagazine.com/c...-life-are-completely-understood/#.VC7uS2d_tgY ]

Another fair warning then re this thread if not you personally: I don't do philosophy. It is easy to show that it is story telling:

P1: "According to my philosophy falsification should work."
P2: "According to my philosophy falsification should not work."

Both perfectly valid philosophies.

The answer has been started to be provided by testing for robustness as per above. (Now to make it quantitative... well, let's wait. At least it is better than the observation that P1 <> P2 => philosophy is BS.)

After inflation ends, the universe has some gravitational potential, some radiant energy and perhaps some matter. Suppose the amount of radiant energy/matter remains constant over the process (does it?). Doesn't the resulting gravitational potential count as energy added by inflation?

I'm not sure I understand. The universe is expanding. so dark energy balances it. (Curvature is roughly zero, so we can discard that GR term as I understand it.)

Then we have those things in common.

Cool. I'm looking forward to see you over at the biology section then! :D

 

[Torbjorn_L]

As long as we don't take these speculative excursions too seriously,

They are not excursions, inflation is 40 years old and has been tested over and over. Not only has it been fruitful, the alternatives has not.

For example, the public is told: "the universe is 5% matter, 20% dark matter and 75% dark energy" as if established fact.

Well, yes, it is observational fact. No one outside the fringe is contesting that, including non-fringe inflation critics.

Besides, it comes out of the CMB spectra (and BAO et cetera), so you don't need inflation for that prediction specifically I think.

 

[Torbjorn_L]

... I don't think that's the case. After all, we do have statistic on such approaches; that's the nature of observational science. We rely heavily on such methodology--Bayesian with a hint of frequentist and a lot of skepticism which is normal and reasonable.

I'm no statistician (though I can use some basics). But I suspect that the above is not quite correct. Bayesian inferences are only promoted to probabilities that all agree on if the bets can be tested (HMM modeled).

Science use frequentist and bayesian probability methods of course, but mainly they use testing (as I described re the measurement theory that underlies it all), so likelihoods:

“To avoid the introduction of prior probabilities, physicists are usually satisfied with
the information contained in the likelihood function. In most cases the MLE and
the likelihood ratio error interval are sufficient to summarize the result. Contrary
to the frequentist confidence interval this concept is compatible with the maximum
likelihood point estimation as well as with the likelihood ratio comparison of discrete
hypotheses and allows to combine results in a consistent way.”

[ http://www-library.desy.de/preparch/books/vstatmp_engl.pdf ]

 

[bapowell]

Science use frequentist and bayesian probability methods of course, but mainly they use testing (as I described re the measurement theory that underlies it all), so likelihoods:

Modern cosmology is done almost exclusively in the Bayesian spirit, both for model selection and parameter estimation.

“To avoid the introduction of prior probabilities

The desire to avoid prior probabilities is a popular criticism levied by frequentists and others who don't understand how inference is done. There is nothing wrong with an inference depending inherently on prior knowledge of the thing being measured; in fact, I'd argue that every statistical inference made in science is based on prior assumptions.

 

[Torbjorn_L]

The desire to avoid prior probabilities is a popular criticism levied by frequentists and others who don't understand how inference is done.

As I described above, "Bayesian inferences are only promoted to probabilities that all agree on if the bets can be tested (HMM modeled)." The problem isn't of understanding (as I understand it), it is the consensus acceptance of results.

in fact, I'd argue that every statistical inference made in science is based on prior assumptions.

I think the reference (which I haven't read yet, but checks with my current understanding of hypothesis testing to be something else than frequentist/bayesian) speaks for itself. Including this as an "others" source, laying out the use of frequentist and bayesian statistics but explicitly rejected that priors is used.

 

[julcab12]

I'm no statistician (though I can use some basics). But I suspect that the above is not quite correct. Bayesian inferences are only promoted to probabilities that all agree on if the bets can be tested (HMM modeled).

Science use frequentist and bayesian probability methods of course, but mainly they use testing (as I described re the measurement theory that underlies it all), so likelihoods:

[ http://www-library.desy.de/preparch/books/vstatmp_engl.pdf ]

... In cosmology, Experimental means are quite limited. We cannot directly create, access,manipulate and infer a scenario with galactic bodies to extrapolate a better testing of the mechanics unlike local scientific methodology. Furthermore, regions in the universe are practically inaccessible by observation. So the only reliable approached is treat the dynamics--theory as an idealized and abstract mathematical model and materialize it by fitting the data using method of analysis and statistical techniques or approximation by prior assumption(like the cosmological principle) in the hope that it will be consistent all throughout. Or whether it can be tested through observations.

My point is, due to the limits of testing. We need an alternative reliable method in the Bayesian and statistical regime/spirit and clues in High energy physics to create an approximate judgement.

 

[Chronos]

I'm not aware of any statistical method that is not axiomatic [i.e., does not rely on prior assumptions].

 

[Torbjorn_L]

I'm not aware of any statistical method that is not axiomatic [i.e., does not rely on prior assumptions].

That method would be hypothesis testing. I'm not even sure if it is axiomatized after the fact, since people still argue about the old compromise that Fisher et al did when putting it on more solid ground.

 

[CKH]

Torbjorn_L,

The reason I call inflation speculative is that the hypotheses (which themselves are loosely defined since there are so many versions) are untested even though they may give the desired result. The level of validity is that a hypothetical explanation works, but the assumptions are untested. This is just the first step in forming a theory. My understanding of Bayesian reasoning is very weak, but it's hard to see how you can assign any measure of certainty at this stage.

 

[bapowell]

The reason I call inflation speculative is that the hypotheses (which themselves are loosely defined since there are so many versions) are untested even though they may give the desired result. The level of validity is that a hypothetical explanation works, but the assumptions are untested.

What are the assumptions that you think are untested?

My understanding of Bayesian reasoning is very weak, but it's hard to see how you can assign any measure of certainty at this stage.

Bayesian reasoning enables one to form a probability distribution over the model space of inflation. Given the data, we select the model that maximizes this probability distribution. What is missing "at this stage"? What else do you think is needed?

 

[CKH]

Many models can predict the same things. E.g. to use an extreme, the initial parameters may have been perfectly fine-tuned in the first place, perhaps due to some other mechanism. It could well be the whim of creation, if there is such a thing. How can you limit the possibilities of creation? (Without cause, it is outside the realm of physics.) Possibly some undiscovered self-balancing action prevents deviation from flatness. Maybe the universe never had the ultra-high density and inhomogeneity assumed, so there is no need for inflation. The need arises from an extrapolation of expansion backwards far beyond the point at which we have any direct evidence of expansion (more than 60 orders of magnitude beyond).

Inflation is one of a variety of models one can conjecture to explain things, limited only by imagination and getting the desired end result. Clearly some people think that this particular extension of physics (to the extent that it is a particular extension) is the right one.

The assumptions I'm aware of are: some as yet unknown field arose at about 10^-36 seconds post singularity, acted for a very short time and blew up the universe at an exponential rate by a factor of 10^60. This scalar(?) field arose and decayed in just the right manner to flatten out (almost?) whatever wrinkles the universe may have had at the point where known physics surely breaks down.

Bayesian reasoning might help if you knew a priori 1) something special must have happened to flatten the universe and 2) that you know what all possible models are and 3) you can somehow suggest which models are more probable.

Of course what you actually referred to was the "model space of inflation", a much narrower range of theories. In that space maybe there is some particular model or models that are preferred, but none has been settled on as correct.

If you think that we are forced to these conclusions (that in general an inflation model is required) by what we know, without untested assumptions, well that's not my understanding of the situation, but I could be wrong.

To be honest, now that I have a better idea of what motivates inflation, whether it is true or not seems to make little difference unless it predicts something new. It would be important if it changes the way we interpret observations of the universe post-CMB. I really don't understand what makes this theory so dear to cosmologist, but apparently they are quite worried about something(s) in BBT to bother proposing it in the first place.

In laboratory science, you don't take theories proposing new physics seriously until experiments demonstrate them (oops, I forgot supersymmetry and string theory). Cosmology is a whole nother ball game because all we have are passive observations.

Until some controlled experiments were done, the theory of spontaneous generation was thought correct.

 

[Drakkith]

Closed, pending moderation.