Sectional curvatures distribution of complexity geometry

• URL: http://arxiv.org/abs/2108.11621v3
• Date: Sun, 21 Aug 2022 02:15:48 GMT
• Title: Sectional curvatures distribution of complexity geometry
• Authors: Qi-Feng Wu
• Abstract summary: In the geometric approach to define complexity, operator complexity is defined as the distance on the operator space. The typical sectional curvatures of this complexity geometry are positive. It can also be shown that in the Hilbert space, 2-surfaces generated by operators of size much smaller than the typical size acting on typical states also have negative curvatures.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In the geometric approach to define complexity, operator complexity is defined as the distance on the operator space. In this paper, based on the analogy with the circuit complexity, the operator size is adopted as the metric of the operator space where path length is the complexity. The typical sectional curvatures of this complexity geometry are positive. It is further proved that the typical sectional curvatures are always positive if the metric is an arbitrary function of operator size. While complexity geometry is usually expected to be defined on negatively curved manifolds. By analyzing the sectional curvatures distribution for $N$-qubit system, it is shown that surfaces generated by Hamiltonians of size smaller than the typical size can have negative curvatures. In the large $N$ limit, the form of complexity metric is uniquely constrained up to constant corrections if we require sectional curvatures are of order $1/N^2$ . With the knowledge of states, the operator size should be modified due to the redundant action of operators, thus is generalized to be state-dependent. Then we use this state-dependent operator size as the metric of the Hilbert space to define state complexity. It can also be shown that in the Hilbert space, 2-surfaces generated by operators of size much smaller than the typical size acting on typical states also have negative curvatures.

Related papers

• Geometric structure and transversal logic of quantum Reed-Muller codes [51.11215560140181]
  In this paper, we aim to characterize the gates of quantum Reed-Muller (RM) codes by exploiting the well-studied properties of their classical counterparts. A set of stabilizer generators for a RM code can be described via $X$ and $Z$ operators acting on subcubes of particular dimensions.
  arXiv  Detail & Related papers  (2024-10-10T04:07:24Z)
• Learning with Norm Constrained, Over-parameterized, Two-layer Neural Networks [54.177130905659155]
  Recent studies show that a reproducing kernel Hilbert space (RKHS) is not a suitable space to model functions by neural networks. In this paper, we study a suitable function space for over- parameterized two-layer neural networks with bounded norms.
  arXiv  Detail & Related papers  (2024-04-29T15:04:07Z)
• Construction of Pseudo-hermitian matrices describing systems with balanced loss-gain [0.0]
  We present a general construction of pseudo-hermitian matrices in an arbitrary large, but finite dimensional vector space. The positive-definite metric which ensures reality of the entire spectra of a pseudo-hermitian operator is presented. We apply the results to construct a generic pseudo-hermitian lattice model of size N with balanced loss-gain.
  arXiv  Detail & Related papers  (2024-01-02T09:45:12Z)
• The Complexity of Being Entangled [0.0]
  Nielsen's approach to quantum state complexity relates the minimal number of quantum gates required to prepare a state to the length of geodesics computed with a certain norm on the manifold of unitary transformations. For a bipartite system, we investigate binding complexity, which corresponds to norms in which gates acting on a single subsystem are free of cost.
  arXiv  Detail & Related papers  (2023-11-07T19:00:02Z)
• Convergence bounds for nonlinear least squares for tensor recovery [0.0]
  We consider the problem of approximating a function in general nonlinear subsets of L2 when only a weighted Monte Carlo estimate of the L2-norm can be computed. By restricting the model class to a neighbourhood of the best approximation, we can derive improved worst-case bounds for the sample complexity.
  arXiv  Detail & Related papers  (2022-08-23T13:25:30Z)
• Threshold size for the emergence of a classical-like behaviour [68.8204255655161]
  We design a procedure to estimate the minimum size beyond which a system is amenable to a classical-like description. The specific case of a magnetic system is considered, with details of a gedanken experiment presented and thoroughly commented.
  arXiv  Detail & Related papers  (2022-03-25T11:31:14Z)
• Bounds on quantum evolution complexity via lattice cryptography [0.0]
  We address the difference between integrable and chaotic motion in quantum theory as manifested by the complexity of the corresponding evolution operators. Complexity is understood here as the shortest geodesic distance between the time-dependent evolution operator and the origin within the group of unitaries.
  arXiv  Detail & Related papers  (2022-02-28T16:20:10Z)
• Coordinate Independent Convolutional Networks -- Isometry and Gauge Equivariant Convolutions on Riemannian Manifolds [70.32518963244466]
  A major complication in comparison to flat spaces is that it is unclear in which alignment a convolution kernel should be applied on a manifold. We argue that the particular choice of coordinatization should not affect a network's inference -- it should be coordinate independent. A simultaneous demand for coordinate independence and weight sharing is shown to result in a requirement on the network to be equivariant.
  arXiv  Detail & Related papers  (2021-06-10T19:54:19Z)
• Geometry of quantum complexity [0.0]
  Computational complexity is a new quantum information concept that may play an important role in holography. We consider quantum computational complexity for $n$ qubits using Nielsen's geometrical approach.
  arXiv  Detail & Related papers  (2020-11-15T18:41:19Z)
• Manifold Learning via Manifold Deflation [105.7418091051558]
  dimensionality reduction methods provide a valuable means to visualize and interpret high-dimensional data. Many popular methods can fail dramatically, even on simple two-dimensional Manifolds. This paper presents an embedding method for a novel, incremental tangent space estimator that incorporates global structure as coordinates. Empirically, we show our algorithm recovers novel and interesting embeddings on real-world and synthetic datasets.
  arXiv  Detail & Related papers  (2020-07-07T10:04:28Z)
• Convex Geometry and Duality of Over-parameterized Neural Networks [70.15611146583068]
  We develop a convex analytic approach to analyze finite width two-layer ReLU networks. We show that an optimal solution to the regularized training problem can be characterized as extreme points of a convex set. In higher dimensions, we show that the training problem can be cast as a finite dimensional convex problem with infinitely many constraints.
  arXiv  Detail & Related papers  (2020-02-25T23:05:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.