Zane Rossi

@henryquantum I could see it being helpful (depending on how the class is structured) to place Simon's, Shor's, and other hidden subgroup problems on equal footing (showing a lot of quantum algorithms are Fourier sampling in disguise). But for QPE it's definitely not the shortest path!

Is there any reason to teach QFT at all in an intro to quantum computing class? From Patrick Rall's paper on phase estimation, it seems potentially superfluous (arxiv.org/pdf/2103.09717).

Tomorrow I am teaching quantum phase estimation in my Intro to Quantum Computing Class for the seventh time. I was prepared to teach it the standard, textbook, Nielsen and Chuang way: applied controlled unitaries and their powers thereof, apply inverse QFT to the ancillas.

We've greatly expanded documentation for v0.2.0 with new examples and discussion! If you're interested in researching (or just learning about) QSP, we hope this package suits all your quantum algorithmic and numerical linear algebraic needs! Find us on PyPI and Github. 😎

Attention quantum signal processing (QSP) ⚛️ and Python 🐍 enjoyers: emerging from radio-silence to say that the pyQSP package 📦 (github.com/ichuang/pyqsp) has been overhauled to greatly improve numerical stability, incorporate better phase-finding methods, and be easier to use!

Excited to share work (joint with @jack_ceroni this summer @MIT) in casting QSP/QSVT as linkable modules achieving function-first, flexible, and efficient quantum algorithms. Check out our paper scirate.com/arxiv/2309.166… or its companion codebase! github.com/ichuang/pyqsp/…

We also describe photonic devices in coupling regimes not formally amenable to closed-form analysis, and discuss the application of QSP-like methods to other Lie groups, with reasons for both positivity and concern. (9/9)

Nevertheless, we show that a large class of useful functions can still be approximated when moving to SU(1,1), some with character fundamentally not possible in the SU(2) setting. (8/9)

Happy to finally discuss two recent projects in fundamental quantum algorithms out now on arXiv, the first from here @MIT and the second with collaborators at NTT BRL in Japan! arxiv.org/abs/2304.14392 and arxiv.org/abs/2304.14383 (1/9).

What looks to be a beautiful, streamlined addition to the required reading for those interested in quantum singular value transformation (QSVT), with ~bonus~ insight on Jordan's lemma and bounded approximation: arxiv.org/abs/2302.14324, from @ewintang and @kevinjtian.

Looking forward to watching presentations and wandering posters at @qip2023; if you’re interested in new work from @MIT on extending QSP & QSVT to the multivariable setting (with much to love for algebraic geometers…) stop by our poster (116) at 7pm today!

The theory of multivariable polynomials is pretty arcane, with far fewer guarantees; nonetheless achievable transformations for M-QSP are well described. M-QSP permits certain speedups incommensurate with those of other quantum algorithms!

QSVT performs univariable polynomial transformations on the singular values of large matrices, and derives great utility; but what if we want multivariable transformations? Can we compute certain joint properties of multiple oracles cheaply? Yes!

Introducing a new quantum algorithm: multivariable quantum signal processing (M-QSP)! Drawing on the versatility of QSVT, we consider the problem of computing multivariable polynomial functions of the singular values of multiple embedded linear operators. arxiv.org/abs/2205.06261

Happy to help with this tutorial for @PRX_Quantum, and hopefully further demystify the (many!) use-cases for the wonderful work by Gilyén, Su, Low, and Wiebe in QSVT (with plenty of examples and diagrams)!

Excited to share our new paper “A Grand Unification of Quantum Algorithms”! We survey the quantum singular value transformation and showcase how it encompasses the three major quantum algorithms (search, simulation, and phase estimation): arxiv.org/abs/2105.02859