Where I would like to be right now
youtube.com/watch?v=stCAaH…
| ||||||||
28 June (2023)
Replying to @AldoSterone111
Merci pour cet exemple qui se déroule "bien" en la circonstance. L'État, la justice, doivent maintenant prendre le relai. Décorer et honorer ce policier courageux, d'abord; retrouver le délinquant et le punir, ensuite. L'avoir abattu–même pour ça–aurait fait du héros un criminel.
27 June (2023)
Replying to @polyquantique
That's more than a "Good question" and I have too many answers that average to "I don't know". But we're thinking about this. Let me know if you have an answer.
26 June (2023)
𝗧𝗵𝗲 𝗠𝗼𝗻𝘁𝘆 𝗛𝗮𝗹𝗹 𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝗶𝗻 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗰𝘀
Part II: Removing photons from a coherent state — where you meet the Minotaur.
(Part I here: )
Part II: Removing photons from a coherent state — where you meet the Minotaur.
(Part I here: )
Fabrice Laussy @fplaussy · 24 Jun 2023
𝗧𝗵𝗲 𝗠𝗼𝗻𝘁𝘆 𝗛𝗮𝗹𝗹 𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝗶𝗻 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗰𝘀
Part I: What people know already — follow Ariadne's thread into the Twitter maze.
Part I: What people know already — follow Ariadne's thread into the Twitter maze.
Imagine a stream of photons that does not suffer from the fundamental weak turnstile limitation of letting some photons leak together, and enforce an actual, strict one-at-a-time rule, of the type one finds nowadays in our dystopian societies (e.g., in airports).
That is, imagine we enforce (never mind how for now) a time gap tɢ so that every time some photons are closer than tɢ, you strip off those that are in the no-photon's land:
Equivalently, instead of suppressing photons, memorylessness of Poisson point processes means that you can understand it as inserting a rigid spacer after each photon.
Because the original stream is, as the simplest and most general case, uncorrelated (Poissonian), which describes a coherent state, we propose to call this hypothetical signal the "𝐺𝑎𝑝𝑝𝑒𝑑 𝑐𝑜ℎ𝑒𝑟𝑒𝑛𝑡 𝑠𝑡𝑎𝑡𝑒".
Now what are the properties of this state? Remember, g⁽²⁾(τ) does not care whether there are photons in between any two separated by τ, only how pairs τ-apart compare to a stream of uncorrelated photons. So you might, maybe, expect something like this:
This is, in some respects, reminiscent of the Monty Hall paradox: once more, g⁽²⁾(τ) correlates all photons pairwise, regardless of whether others are in between.
It may be surprising, therefore, that photons initially uncorrelated are now expected to be found together, as a result of the knowledge (although a useless one for g⁽²⁾) that no photons are in between.
In the original stream, there might have been photons, or not, this does not affect the two τ-apart photons. If I tell you, however, for sure there are no photons, then this changes everything, at all times!
Bunching also evokes photons clumping together. It may also be surprising, therefore, that removing photons results in those remaining being "closer together", as opposed to "farther apart".
Like Marylin's insight, this is actually easy to understand. Sana Khalid and I @wlv_uni wrote a paper now on arXiv arxiv.org/abs/2306.13646. What is interesting is that this toy model actually says a lot about the fundamental nature of what it is to be a single-photon source.
If you ask Aspect🏅, for instance, he will tell you that it “is not enough to have single-photon wavepackets to have single-photon light” but that one “need[s] in addition to know at which time each single-photon wavepacket is emitted” [Exp. Methods Phys. Sci. 45, 315 (2013)]
Most experts would vouch that pulsed emission makes best SPS. There is some truth in that: here is our gapped coherent state when it is bright (lots of photons). While this is a stationary source—g⁽²⁾(∞)=1—if you show it to anybody they will tell you "oh, that's a pulsed SPS".
Really smart and perceptive people like @JCLoredoR would however catch at a glance from a random picture on social media that there is something uncanny and ask you what it is:
Fabrice Laussy @fplaussy · 27 May 2023
A special day yesterday as we celebrated the Physics that took place in Wolverhampton for the last 7 years. Here explaining a peculiar type of oscillation that was one of our iconic discoveries.
It is not pulsed: it is a CW perfect single-photon source; perfect in the sense that it does not allow ever two photons to be found together, unlike the conventional 2PS (incoherently pumped two-level systems [2LS]) that only makes this less likely.
We also understand better why the incoherent 2LS is not good: it merely inserts a gap that is not rigid but also Poisson distributed so if your photons are close (Poisson bursts) and their repulsion is small (Poisson distribution peaks at τ=0), then you still have a photon clump.
The inspiration for this work was our realization that genuine sub-natural linewidth antibunched sources feature a plateau in g⁽²⁾(τ) [Quantum Sci. Technol. 3:045001, 2018 https://t.co/mHUB8eEVjX]
This source appears to have many wondrous properties, one of them being that it provides much better single-photons (purity) than other schemes. I believe people should look into it more carefully.
We do not know yet exactly how the coherent fraction (wave) modulates the photons (particles) to produce such a plateau but we have clues [PRL 125:170402 2020 dx.doi.org/10.1103/PhysRe… ; Phys. Scr. 98:055104 2023 https://t.co/ldWYrbntzo]
This confirms that the turnstile repulsion of the standard 2LS is not good enough. Look how g⁽²⁾(τ) looks like for a repulsion that is not the feeble Poisson push but a stronger—zero probability of zero delay—Maxwell distribution. Now you see genuine single-photons! No noise:
By the way, the oscillations come from the bright signal. The gap-only version (a plateau, no oscillations) is possible for dim gapped-coherent states:
And guess what: the sub-natural linewidth antibunching, when it only shows the plateau, also has a very weak signal: one does not remove photons in a gap but, instead, we remove photons that are coherent. Somehow, these appear to be related.
Doesn't it tell you a thing or two about single-photon emission? But we have completed our journey back from the maze with this 25th tweet. I let you ponder that naked 2LS don't address photon piggybacking well enough and that fundamental alternatives may be needed.
25 June (2023)
Playing with Google Scholar laussy.org/wiki/Blog:Hack… - That's by how much, for instance, Heisenberg gets more cited than Schrödinger: the trend is slightly reversing.
Julia's poem turned into a micro-musical (that she also composed)— her sister, who provides the special effects in the domain she masters most—terror—already knows it by heart youtube.com/shorts/mxByRX-… youtube.com/watch?v=uVkUK1…
24 June (2023)
𝗧𝗵𝗲 𝗠𝗼𝗻𝘁𝘆 𝗛𝗮𝗹𝗹 𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝗶𝗻 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗰𝘀
Part I: What people know already — follow Ariadne's thread into the Twitter maze.
Part I: What people know already — follow Ariadne's thread into the Twitter maze.
The Monty Hall problem is a famous problem in probability theory, named after the host of a TV show (who was playing the game in a different way) whereby you are offered to open one out of three doors.
One of the doors hides a valuable prize (say, a car), the two others something of little value (in the older days of this puzzle, people were putting a goat as a symbol of insignificance!) The host knows where is the car but you don't.
After you called a door, the host opens one of the doors you did not choose, to reveal a goat (remember, he knows where is the car and he also assumes that's what you want, not the goat). He can always do that whatever your initial choice.
Most people nowadays know the answer. If you don't, here is your chance to think about it: what would you do (to win the car)?
Since the host can always open one of the losing doors, it would appear that he revealed no information, and that you can keep or swap the door with the same, now 1/2, probability to win the car (1/3 previously).
It is counter-intuitive to most (at least it was to Erdös) that you should switch the door, thereby increasing your odds to 2/3.
One of the fascinating aspects of this puzzle is how it was straightforwardly solved as if a trifle by the cleverest woman (highest IQ) who was tackling all sorts of problems in a column ("ask Marilyn") and how this generated a shitstorm (pardon the term but that's what it was).
Vos Savant (what a name) in her very first and concise reply already clarified the essence of the paradox. Here is in fact her full, now legendary, column. What a genius indeed. She didn't need long Twitter threads to make her point!
It took several "Referee" rounds with the public that involved delightful and instructive arguments (a small selection below) before she could finally convince everybody with a pioneering (1990) delocalized, social, Monte Carlo experiment of readers simulating the game at home.
Interestingly, Marylin is known to also have made quite basic mistakes on simpler, school book problems. Look at this one: "I manage a drug-testing program for an organization with 400 employees. Every three months, a random-number generator selects 100 names for testing. [...]
Afterward, these names go back into the selection pool. Obviously, the probability of an employee being chosen in one quarter is 25 percent. But what is the likelihood of being chosen over the course of a year?" — she answered it's always 25%
The correct answer is 1−0.75⁴≈68%. Someone still dared to correct her, but she did, however, recognized her mistake immediately. She was still attending her column recently and remains inspiring from the start even on things that happen at the very moment. I love you, Marylin!
Now let's turn to a similar paradox for people familiar with photons and their correlations, the so-called quantum-opticians, who quantify the density of two-photon events in a stream of observations with the Nobel-prize winning quantity g⁽²⁾(τ).
This describes the probability of two photons being separated by the time delay τ. Importantly, this ignores whether other photons are in between them or not: it compares how likely are two photons to be found a delay τ apart as compared to if they would be distributed by chance.
For a stream of uncorrelated photons, g⁽²⁾(τ)=1, since any two-photons found are, precisely, there by chance only, with no correlation, reason or connection to the other photons. This describes photons found in the light of a laser.
Quantum optics itself started with the insight of Hanbury Brown that photons from natural light (like the sun) are correlated: they tend to be detected together. This results in coincidences (at τ=0) that double the uncorrelated case g⁽²⁾(0)=2. This was already paradoxical.
Here I show you the only picture I could ever find of Hanbury Brown with the guy who helped him work out the maths of his fantastic insight. I show this because even if you know their names, it's probably the first time you see them together.
The favorite type of quantum light for quantum opticians is still another one, from the so-called single-photon source (SPS), in which photons do not come together or randomly, but separately, i.e., one at a time, in particular, there should never be more than one of them.
People usually imagine this as the result of a two-level system (like an atom) acting as a sieve from their ground state/excited state (on/off) character, blocking uncorrelated photons from a laser in a way similar to how a turnstile would regulate people in a crowd.
The image is correct at least in the sense that turnstiles are not immune to so-called piggybacking (two in a row) and tailgating (the same but without consent) and other practices one can witness or practice in the metro.
This results in what is a fundamental limitation of such two-photon emission: it does suppress to some extent—quantitatively—but not completely (qualitatively) multiple photons. The typical g⁽²⁾(τ) of these sources looks like this: the full prohibition is at τ=0 only.
I et al. have discussed at length and over the years how this, at the fundamental photodetection level, hinders a perfect single-photon source: see PRL 109:183601 (2012), arXiv:1610.06126, 94:063826 (2016), Sci. Rep. 8:6975 (2018), PRA 105:023724 (2022)... doi.org/10.1103/PhysRe…
But there's no more thread to pull from, in the Twitter maze (25 tweets max) so I see you on Monday for Part II: "Removing photons from a coherent state — where you meet the Minotaur".
Part II here:
Fabrice Laussy @fplaussy · 26 Jun 2023
𝗧𝗵𝗲 𝗠𝗼𝗻𝘁𝘆 𝗛𝗮𝗹𝗹 𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝗶𝗻 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗰𝘀
Part II: Removing photons from a coherent state — where you meet the Minotaur.
(Part I here: )
Part II: Removing photons from a coherent state — where you meet the Minotaur.
(Part I here: )
23 June (2023)
A teaser for the week-end and an early clue for @JCLoredoR (full answer on Monday): Fig. 4.
22 June (2023)
magical
mum
wonderful
wife
growing
girl
nothing will never be enough to celebrate you, our Luz bringer. May the cuarentona years be less rock & roll youtube.com/watch?v=5_ucNN… and mas de dar vueltas youtube.com/watch?v=stCAaH…, Señora.
mum
wonderful
wife
growing
girl
nothing will never be enough to celebrate you, our Luz bringer. May the cuarentona years be less rock & roll youtube.com/watch?v=5_ucNN… and mas de dar vueltas youtube.com/watch?v=stCAaH…, Señora.
15 June (2023)
Replying to @JCLoredoR
you will be the first to know.
9 June (2023)
He seems to have also enjoyed the talk himself from the audience!
8 June (2023)
@goutraillon On y revient toujours. Il semble avoir déjà tout vu, à une époque où ça ne faisait que commencer. À chaque relecture j'ai l'impression que c'est la première fois, qu'il est revenu sur le texte de ma mémoire pour le mettre en adéquation avec ce qui s'est passé depuis.
@goutraillon Bernanos c'est comme un apôtre des temps modernes.
ah, it seems the Web of Science finally found me. I am always appalled at how bad all these "metrics" things really are in the first place, let alone what they mean, they're not well compiled in the first place. Speak of a digital age!
