Timo Hillmann

find me over at the alternative: @timohillmann.bsky.social I'm out of here.

@letonyo For any n=m you should be able to rewrite this as n+m derivative wrt to a. Then the sum is just over the Gaussian part with a global prefactor of 1/a^2 if I did it right in my head.

update, for m=n=0, one (well, Mathematica) gets EllipticTheta[3, 0, exp(-a^2)]^2

@eshaswaroop @mike_vasmer anything that has meta-checks that scale with the code, really. The four-fold HGP based on the [25s^2, s^2] family of arxiv.org/abs/2109.14609 should do for sure. One could also construct explicit meta-checks for the code obtained from the standard HGP and check for that property.

@mike_vasmer congrats on the neat results Mike and @timohillmann ! Curious which families of non-topological codes could you get FT with O(1) rounds of syndrome measurements?

Check out @mike_vasmer views on our project about fault complexes! Thanks for keeping an open mind about my vague ideas and intuitions throughout!

@jenseisert @XanaduAI Thanks, Jens! This has indeed been a result that changed my understanding of quantum error correction in space-time!

@timohillmann @XanaduAI This is really cool, Timo.

Last year I’ve spent some months at @XanaduAI. I wanted to use that time to finally properly understand fault-tolerant MBQC. Digging in the literature I found that previous works didn’t fit my style of thinking. It was time for something new! scirate.com/arxiv/2410.129…

Thanks again to the whole @XanaduAI team and especially my close collaborators on this project, Guillaume, Ilan, and @mike_vasmer

I hope that the concept of the fault complex proves to be useful beyond our work. We’ve sketched out a few extensions to subsystem codes and also circuit-level noise models. I see it as a step of going away from the picture of static codes to fault-tolerant channels!

@MWalschaers The Moyal product is something I deeply remember from the days of my bachelor project! We then discovered that (numerically) the Wigner function was not stable enough, and I had to learn the Voros product as well. Fun times! Looking forward to having a look at your introduction!

@letonyo @nic_delfosse I agree. I have noticed though that their are parts of the (very abstract/theoretical) community that don't value rep code experiments. I think it is a great way to gain confidence in effective noise models and discover effects that one initially forgot about, or new physics

@nic_delfosse it'd be great to know whether similar error floors occur for other platforms (e.g. neutral atoms, ions).

Great QEC experiment at Google. The surface code below threshold, which is already great, but they also reach a very low error floor with the repetition code at 10^-10.

@decodoku @CraigGidney Could they by any chance mean this? “Before” would be very loosely defined in that case of course. arxiv.org/abs/2406.12267

Thanks to the team for this fun collaboration; @lucasberent @armanda_oq @jenseisert @rbrtwll @QuantumGizmos (7/7)

Our recipe is openly available as part of the LDPCv2 package which contains lots of other good stuff. stim/sinter integration, overlapping window decoders for circuit-level noise, you name it. github.com/quantumgizmos/… (6/7)

Tired of quantum error correction feeling like a bad trip? Fear no more! Meet our new method: Localized Statistics Decoding (LSD). Today's the day that we bring it to the masses! scirate.com/arxiv/2406.186… (1/7)