En 1976, Connes demande si toute algèbre de von Neumann finie se plonge dans un ultraproduit d'algèbres de matrices. En 1980, Tsirelson demande si, dans la formulation mathématique de la mécanique quantique, autoriser des espaces de Hilbert de dimension infinie change fondamentalement le modèle. En 1993, Kirchberg conjecture que le produit tensoriel de deux copies de la C*-algèbre pleine du groupe libre de rang infini dénombrable peut être muni d'une unique norme de C*-algèbre. De manière surprenante et non triviale, ces trois problèmes sont en fait équivalents, c'est maintenant bien compris. Ces problèmes viennent d'être résolus, par la négative, avec des méthodes d'informatique : calculabilité, complexité, et informatique quantique. Je ferai de mon mieux pour raconter les grandes lignes de cette très longue preuve.
We prove quantum soundness of the total-degree diagonal line-vs-point test using the individual-degree soundness theorem of Ji, Natarajan, Vidick, Wright, and Yuen. A random change of coordinates yields projective polynomial decoders of total degree at most
md. The uniform-line slice of the test bounds the weight of outcomes of degree greater than
d, which are removed by a common relabeling. This reduction does not yield a dimension-independent soundness bound: the
poly(m) dependence of the individual-degree theorem persists, as discussed in Section 1.2 of arXiv:2009.12982.
We show that a partition of the unity (or POVM) on a Hilbert space that is almost orthogonal is close to an orthogonal POVM in the same von Neumann algebra. This generalizes to infinite dimension previous results in matrix algebras by Kempe–Vidick and Ji–Natarajan–Vidick–Wright–Yuen. Quantitatively, our result are also finer, as we obtain a linear dependance, which is optimal. We also generalize to infinite dimension a duality result between POVMs and minimal majorants of finite subsets in the predual of a von Neumann algebra.
We investigate the connection between the complexity of nonlocal games and the arithmetical hierarchy, a classification of languages according to the complexity of arithmetical formulas defining them. It was recently shown by Ji, Natarajan, Vidick, Wright and Yuen that deciding whether the (finite-dimensional) quantum value of a nonlocal game is 1 or at most 1/2 is complete for the class Σ1 (i.e., ). A result of Slofstra implies that deciding whether the commuting operator value of a nonlocal game is equal to 1 is complete for the class Π1 (i.e., coRE).