Christopher K. Long
@christopher-k-long.bsky.social
4th year PhD Student at the University of Cambridge focusing on Quantum Computing
(he/him)
Google Scholar: https://scholar.google.com/citations?user=GRSIcsEAAAAJ
GitHub: https://github.com/Christopher-K-Long
ORCiD: https://orcid.org/0009-0001-3230-942X
(he/him)
Google Scholar: https://scholar.google.com/citations?user=GRSIcsEAAAAJ
GitHub: https://github.com/Christopher-K-Long
ORCiD: https://orcid.org/0009-0001-3230-942X
For example, numerical quantum optimal control techniques such as GRAPE and CRAB fall within the framework of VQAs and are used ubiquitously today to tune quantum processors.
November 30, 2025 at 12:10 PM
For example, numerical quantum optimal control techniques such as GRAPE and CRAB fall within the framework of VQAs and are used ubiquitously today to tune quantum processors.
My point is I wouldn't declare VQAs doomed—that's too strong. Most no-go theorems have loopholes and assumptions. They tell us where it is still worth looking. I would just say that VQAs' use cases are increasingly constrained, and so there are probably more promising directions.
November 30, 2025 at 12:10 PM
My point is I wouldn't declare VQAs doomed—that's too strong. Most no-go theorems have loopholes and assumptions. They tell us where it is still worth looking. I would just say that VQAs' use cases are increasingly constrained, and so there are probably more promising directions.
Given the drastic improvements in noise tolerance of VQE algorithms since colleagues and I wrote [doi:10.1038/s41534-024-00808-x](doi.org/10.1038/s415...), it may even be possible to use coherently executed shots to improve the runtime of expectation value estimation.
Quantifying the effect of gate errors on variational quantum eigensolvers for quantum chemistry - npj Quantum Information
npj Quantum Information - Quantifying the effect of gate errors on variational quantum eigensolvers for quantum chemistry
doi.org
November 30, 2025 at 12:10 PM
Given the drastic improvements in noise tolerance of VQE algorithms since colleagues and I wrote [doi:10.1038/s41534-024-00808-x](doi.org/10.1038/s415...), it may even be possible to use coherently executed shots to improve the runtime of expectation value estimation.
The large shot count is a problem. I hope that better techniques for estimating expectation values will be developed.
November 30, 2025 at 12:10 PM
The large shot count is a problem. I hope that better techniques for estimating expectation values will be developed.
Given that many alternative methods for eigenvalue estimation require polynomial overlap between the initial state and the eigenstate, it may be that VQAs are still needed for initial-state preparation.
November 30, 2025 at 12:10 PM
Given that many alternative methods for eigenvalue estimation require polynomial overlap between the initial state and the eigenstate, it may be that VQAs are still needed for initial-state preparation.
Guarantees are good but not vital. I think once there is good enough hardware someone will run the best known VQE for a molecule (like FeMoco) and it will either work or it won't. They will probably use tricks like adaptivity, pulse optimization, and error mitigation as asymptotics won't matter.
November 29, 2025 at 1:15 PM
Guarantees are good but not vital. I think once there is good enough hardware someone will run the best known VQE for a molecule (like FeMoco) and it will either work or it won't. They will probably use tricks like adaptivity, pulse optimization, and error mitigation as asymptotics won't matter.
Thank you for the reference, I will add it to the read list
November 29, 2025 at 1:09 PM
Thank you for the reference, I will add it to the read list
I agree with this for non-adaptive VQAs. Last I checked, no one had proved barren plateaus cause problems for adaptive VQAs—[numerical evidence](doi.org/10.1038/s415...) suggests the contrary. There is still some life in VQAs. Nonetheless, I'm glad to have diversified my research away from VQAs.
Adaptive, problem-tailored variational quantum eigensolver mitigates rough parameter landscapes and barren plateaus - npj Quantum Information
npj Quantum Information - Adaptive, problem-tailored variational quantum eigensolver mitigates rough parameter landscapes and barren plateaus
doi.org
November 29, 2025 at 9:51 AM
I agree with this for non-adaptive VQAs. Last I checked, no one had proved barren plateaus cause problems for adaptive VQAs—[numerical evidence](doi.org/10.1038/s415...) suggests the contrary. There is still some life in VQAs. Nonetheless, I'm glad to have diversified my research away from VQAs.
Wish I could have gone, looking forward to #QuiDiQua4!
November 11, 2025 at 7:15 AM
Wish I could have gone, looking forward to #QuiDiQua4!
Thank you for sharing this! Please give us an update of you hear back from Springer Nature.
November 7, 2025 at 6:08 AM
Thank you for sharing this! Please give us an update of you hear back from Springer Nature.
At a glance this looks like it considers higher orders than Fermi's Golden rule so should give a better estimate. I hadn't thought about phonon spin coupling—which seems important in your use case. Phonon spin coupling could also be studied with Fermi's Golden rule. But higher orders are better.
November 3, 2025 at 7:31 PM
At a glance this looks like it considers higher orders than Fermi's Golden rule so should give a better estimate. I hadn't thought about phonon spin coupling—which seems important in your use case. Phonon spin coupling could also be studied with Fermi's Golden rule. But higher orders are better.
I think being hit by a meteroid might be harder to error correct than by a few cosmic rays...
November 3, 2025 at 7:22 PM
I think being hit by a meteroid might be harder to error correct than by a few cosmic rays...
Let me know if this helps :)
October 30, 2025 at 7:02 AM
Let me know if this helps :)
Note due to approximations the rule breaks down for large t. Differentiating this at t=0 will give the initial rate, R, where the approximations are most valid. The if amplitude damping noise is modeled as 1-e^{-t/T1} then differentiating at zero we can fit for T1 and we find T1=1/R.
October 30, 2025 at 7:02 AM
Note due to approximations the rule breaks down for large t. Differentiating this at t=0 will give the initial rate, R, where the approximations are most valid. The if amplitude damping noise is modeled as 1-e^{-t/T1} then differentiating at zero we can fit for T1 and we find T1=1/R.
I don't know anything about diradical molecules, but hopeful the following is helpful:
In the stack I retain the sinc terms. You can either approx these as delta function first or just integrate the probability over the emission frequency. This will give total probability of emission after a time t.
In the stack I retain the sinc terms. You can either approx these as delta function first or just integrate the probability over the emission frequency. This will give total probability of emission after a time t.
October 30, 2025 at 7:02 AM
I don't know anything about diradical molecules, but hopeful the following is helpful:
In the stack I retain the sinc terms. You can either approx these as delta function first or just integrate the probability over the emission frequency. This will give total probability of emission after a time t.
In the stack I retain the sinc terms. You can either approx these as delta function first or just integrate the probability over the emission frequency. This will give total probability of emission after a time t.
It looks like Bluesky doesn't like transparent images. Here is take two:
October 28, 2025 at 5:58 PM
It looks like Bluesky doesn't like transparent images. Here is take two:
"A good description of the phase estimation algorithm can be found in Mosca's Ph.D. thesis"—of which I found a copy here:
www.karlin.mff.cuni.cz/~holub/soubo...
(4/4)
www.karlin.mff.cuni.cz/~holub/soubo...
(4/4)
www.karlin.mff.cuni.cz
October 28, 2025 at 5:53 PM
"A good description of the phase estimation algorithm can be found in Mosca's Ph.D. thesis"—of which I found a copy here:
www.karlin.mff.cuni.cz/~holub/soubo...
(4/4)
www.karlin.mff.cuni.cz/~holub/soubo...
(4/4)
Cleve, Ekert, Macchiavello and Mosca (doi.org/10.1098/rspa...) "integrated several of the techniques of Shor and Kitaev"—section 5 and figure 6 present the algorithm as we know it today.
(3/4)
(3/4)
Quantum algorithms revisited | Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences
Quantum computers use the quantum interference of different computational paths to
enhance correct outcomes and suppress erroneous outcomes of computations. A common
pattern underpinning quantum algor...
doi.org
October 28, 2025 at 5:53 PM
Cleve, Ekert, Macchiavello and Mosca (doi.org/10.1098/rspa...) "integrated several of the techniques of Shor and Kitaev"—section 5 and figure 6 present the algorithm as we know it today.
(3/4)
(3/4)
A quick check in Nielsen and Chuang (Chapter 5, History and further reading, page 246 in my copy):
Kitaev introduced the phase estimation with a single output bit (arxiv.org/abs/quant-ph...).
(2/4)
Kitaev introduced the phase estimation with a single output bit (arxiv.org/abs/quant-ph...).
(2/4)
Quantum measurements and the Abelian Stabilizer Problem
We present a polynomial quantum algorithm for the Abelian stabilizer problem which includes both factoring and the discrete logarithm. Thus we extend famous Shor's results. Our method is based on a procedure for measuring an eigenvalue of a unitary operator. Another application of this procedure is a polynomial quantum Fourier transform algorithm for an arbitrary finite Abelian group. The paper also contains a rather detailed introduction to the theory of quantum computation.
arxiv.org
October 28, 2025 at 5:53 PM
A quick check in Nielsen and Chuang (Chapter 5, History and further reading, page 246 in my copy):
Kitaev introduced the phase estimation with a single output bit (arxiv.org/abs/quant-ph...).
(2/4)
Kitaev introduced the phase estimation with a single output bit (arxiv.org/abs/quant-ph...).
(2/4)
I meant the one with the QFT (diagram from Wikipedia below). I've always heard it called Kitaev's QPEA. In the thread is some history—it looks like many worked on it. The algorithm is HSP with the hiding function given by the sequence of controlled unitaries, and the period being the phase.
(1/4)
(1/4)
October 28, 2025 at 5:53 PM
I meant the one with the QFT (diagram from Wikipedia below). I've always heard it called Kitaev's QPEA. In the thread is some history—it looks like many worked on it. The algorithm is HSP with the hiding function given by the sequence of controlled unitaries, and the period being the phase.
(1/4)
(1/4)
I think seeing the relation between Kitaev's phase estimation algorithm and the hidden subgroup problem algorithm is useful. That said it is definitely good that students know it isn't the only or optimal way to do it.
October 28, 2025 at 7:11 AM
I think seeing the relation between Kitaev's phase estimation algorithm and the hidden subgroup problem algorithm is useful. That said it is definitely good that students know it isn't the only or optimal way to do it.
For T1 you can use Fermi's Golden rule. Here is a post I made a few years ago for absorbing a photon:
physics.stackexchange.com/a/649933/305...
Keep the emission term instead. Also you will probably need more than minimal couping to the EM field to get the spin flip on emission.
physics.stackexchange.com/a/649933/305...
Keep the emission term instead. Also you will probably need more than minimal couping to the EM field to get the spin flip on emission.
What happens to an electron if given quantized energy to jump to a full orbital?
Let's consider the element neon. Its ground-state electron configuration is: $1s^2 2s^2 2p^6$.
What would happen if enough energy was given for one electron in the $1s$ orbital to jump to the $2s$
physics.stackexchange.com
October 25, 2025 at 10:38 AM
For T1 you can use Fermi's Golden rule. Here is a post I made a few years ago for absorbing a photon:
physics.stackexchange.com/a/649933/305...
Keep the emission term instead. Also you will probably need more than minimal couping to the EM field to get the spin flip on emission.
physics.stackexchange.com/a/649933/305...
Keep the emission term instead. Also you will probably need more than minimal couping to the EM field to get the spin flip on emission.
PySTE now has pre-built wheels for Python 3.14!
October 19, 2025 at 10:54 PM
PySTE now has pre-built wheels for Python 3.14!
That makes sense. I would be interested in the slides afterwards is Alex is happy to share them.
October 16, 2025 at 2:33 PM
That makes sense. I would be interested in the slides afterwards is Alex is happy to share them.