On Complexity and Duality
Jeff Murugan, Zayd Pandit, and Hendrik J R Van Zyl · 2024
Read On Complexity and Duality by Jeff Murugan, Zayd Pandit, and Hendrik J R Van Zyl (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2411-02546.txt.
Description
Abstract We explore the relationship between complexity and duality in quantum systems, focusing on how local and non-local operators evolve under time evolution. We find that non-local operators, which are dual to local operators under specific mappings, exhibit behavior that mimics the growth of their local counterparts, particularly when considering state complexity.
Catalog record
| CREATOR | Jeff Murugan, Zayd Pandit, and Hendrik J R Van Zyl |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2411-02546
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On Complexity and Duality Jeff Murugan, Zayd Pandit, and Hendrik J R Van Zyl Abstract We explore the relationship between complexity and duality in quantum systems, focusing on how local and non-local operators evolve under time evolution. We find that non-local operators, which are dual to local operators under specific mappings, exhibit behavior that mimics the growth of their local counterparts, particularly when considering state complexity. For the open transverse Ising model this leads to a neat organisation of the operator dynamics on either side of the duality, both consistent with growth expected in a quadratic fermion model like the Kitaev chain. When examing periodic chains, however, the mapping of boundary terms provides access to multiple branches of highly complex operators. These give rise to much larger saturation values of complexity for parity-mixing operators and are in contrast to what one would expect for a quadratic Hamiltonian. Our results shed light on the intricate relationship between non-locality, complexity growth, and duality in quantum systems. 1 Introduction Understanding how operators evolve and grow in quantum many-body systems has become central to the study of quantum complexity and chaos [ 1 , 2 , 3 , 4 ] . One promising approach to quantify this growth is through Krylov complexity [ 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 ] , which measures how an operator spreads within its Hilbert space under time evolution. Krylov complexity is determined by the Lanczos coefficients, which describe the recursive application of the Hamiltonian on a chosen initial operator [ 4 , 12 ] . This framework provides a powerful lens for investigating dynamical properties such as operator growth, scrambling, and quantum chaos [ 13 , 14 , 15 , 16 , 17 , 8 , 18 , 11 ] . The systematic growth of Krylov complexity reveals distinct signatures in integrable and chaotic systems, with integrable models typically exhibiting polynomial growth, while chaotic systems show faster, often exponential, operator spreading [ 8 , 19 , 20 , 21 ] . In this work, we explore Krylov complexity in the context of theories related under a duality transformation, testing the hypothesis that dual operators exhibit similar growth in Hilbert space, even when one operator is local and the other is non-local. The idea that non-local operators (those dual to local operators through bosonization or fermionization) could exhibit growth patterns akin to local operators is intriguing, particularly in the context of state complexity, which measures the difficulty of preparing quantum states from a reference state [ 10 ] . Our primary goal is to understand how duality affects the operator growth and whether these seemingly distinct operators exhibit comparable behavior in terms of Krylov complexity. The duality we focus on is the transverse field Ising model (TFIM) [ 22 ] and its dual [ 23 ] , the Kitaev chain [ 24 ] , a 1D system of free Majorana fermions. The Jordan-Wigner transformation [ 25 ] , a well-known fermionization map, relates these two models by mapping the spin degrees of freedom in the TFIM to fermionic excitations in the Kitaev chain. In this mapping, local spin operators in the TFIM become highly non-local string operators in the fermionic theory, and vice versa. Despite this difference in locality, both theories share the same spectrum, raising a natural question: do local operators in the Ising model and their non-local duals in the Kitaev chain exhibit similar Krylov complexity? Naively, it would seem that answer to the above question is a simple yes. Any observable may be expressed in terms of the eigenbasis for the respectively models and, since the energies are identical and the eigenstates may be mapped to each other exactly, one may expect the correlators to all match. However, as highlighted in [ 23 , 26 ] , though the models are mathematically equivalent they are physically distinct. The distinction may be boiled down to the presence (absence) of topological order in the Kitaev chain (Ising model) and its robustness under local perturbations [ 23 ] . With this additional knowledge, the question above becomes significantly more subtle. As we will demonstrate, the naive intuition holds for the spread complexity of a time-evolved reference state chosen from one of the parity sectors, but not for the K-complexity of the associated operator in general. The paper is organised as follows. In section 2 we introduce the Kitaev chain as well as the Ising chain and the dual Hamiltonian that it maps to under the Jordan-Wigner transformation. We highlight the distinctions between the two as well as the operators that one may expect to probe this distinction. In section 3 we briefly introduce our chosen diagnostic, Krylov complexity, and the methods we employ to compute it. Section 4 contains the results of our investigation, and we finally conclude with discussion of these and possible future directions in section 5. 2 Boson-Fermion Duality The concrete duality that we will focus on in this paper is the hardcore boson-fermion duality realised by the Jordan-Wigner transformation [ 25 ] . The primary reason for this is that the systems related in this way are finite-dimensional so that the computation of complexity avoids some of the difficulties presented in bosonic systems [ 27 , 28 ] . Furthermore, the specific pair of dual systems we have in mind, the Ising and Kitaev chains, have already been studied in the context of the complexity literature providing a nice way to benchmark our results [ 4 , 29 , 30 , 31 , 32 ] . In this paper, we will specifically be interested in understanding the behavior of K-complexity as viewed through the duality map. As articulated in [ 4 ] , a duality transformation can map certain non-local operators on one side of the duality to local operators on the other side of the duality. As such, the growth of operators may not be correlated with the growth of complexity. Indeed, in integrable models the dynamics of operator growth may be restricted to a symmetry sector and operators classified according to the symmetry sector within which they fall. An (imminently relevant) example of this may be found in the L -site Kitaev chain with general twisted boundary conditions [ 33 ] \displaystyle H_{K,b} \displaystyle= \displaystyle\sum_{i=1}^{L-1}\left[-t(c_{j+1}^{\dagger}c_{j}+c_{j}^{\dagger}c_{j+1})+\Delta(c_{j+1}^{{\dagger}}c_{j}^{\dagger}+c_{j}c_{j+1})\right]-\sum_{j=1}^{L}\mu(c_{j}^{\dagger}c_{j}-c_{j}c_{j}^{\dagger}) (1) \displaystyle+b\left[-t(e^{i\phi_{1}}c_{L}^{\dagger}c_{1}+e^{-i\phi_{1}}c_{1}^{\dagger}c_{L})+\Delta(e^{i\phi_{2}}c_{L}^{\dagger}c_{1}^{\dagger}+e^{-i\phi_{2}}c_{1}c_{L})\right]\;. This Hamiltonian may be recast [ 33 ] as a quadratic Hamiltonian in terms of the Majorana-like generators, \displaystyle\gamma^{a}_{j} \displaystyle= \displaystyle c_{j}+c_{j}^{\dagger} \displaystyle\gamma^{b}_{j} \displaystyle= \displaystyle i(c_{j}^{\dagger}-c_{j})\;. (2) Since it is quadratic, the dynamics of an operator of the form \prod_{i=1}^{L}(\gamma^{a}_{i})^{n_{i}}(\gamma^{b}_{i})^{m_{i}}\ \ \ ;\ \ \ n_{i},m_{i}\in\left\{0,1\right\} (3) is restricted to the subspace of fixed n=\sum_{i}(n_{i}+m_{i})\;, (4) which has dimension \textnormal{dim}=\left(\begin{array}[]{c}2L\\ n\end{array}\right)\;. (5) The interested reader may consult appendix A for further details. A similar decomposition holds for states. Since the Hamiltonian is quadratic in creation / annihilation operators it may be diagonalised by means of a Bogoliubov transformation. The Fock space may be decomposed into subspaces consisting of n Bogoliubov quasi-particles. These subspaces do not mix under time-evolution. In what follows when we make references to subspaces we will refer to the subspace of operators unless stated otherwise. This is important, since these spaces are rather different.