betaFeedback

WorksPhysics and Astronomy

Symmetry from Entanglement Suppression

Abstract

Symmetry is among the most fundamental and powerful concepts in nature, whose existence is usually taken as given, without explanation. We explore whether symmetry can be derived from more fundamental principles from the perspective of quantum information. Starting with a two-qubit system, we show there are only two minimally entangling logic gates: the Identity and the SWAP, where SWAP interchanges the two states of the qubits. We further demonstrate that, when viewed as an entanglement operator in the spin-space, the SS-matrix in the two-body scattering of fermions in the ss-wave channel is uniquely determined by unitarity and rotational invariance to be a linear combination of the Identity and the SWAP. Realizing a minimally entangling SS-matrix would give rise to global symmetries, as exemplified in Wigner's spin-flavor symmetry and Schrödinger's conformal invariance in low energy Quantum Chromodynamics. For NqN_q species of qubit, the Identity gate is associated with an [SU(2)]Nq[SU(2)]^{N_q} symmetry, which is enlarged to SU(2Nq)SU(2N_q) when there is a species-universal coupling constant.

Introduction

Symmetry plays an essential role with respect to the laws of nature. Mathematically defined as invariance under a specified group of transformations, the notion of symmetry has provided unity to vastly different physical phenomena. For example, different physical systems with the same underlying symmetries have the same critical exponents at second order phase transitions. The existence of conserved quantities, such as energy and momentum, is typically a consequence of an underlying symmetry, in this case invariance under time and spatial translations. Effective field theories, including the Standard Model, are defined by writing the most general Lagrangian consistent with known symmetries.

But what is the origin of symmetry? In a physical theory symmetries are usually given as inputs at short distances and are valid over a large range of length scales. Sometimes symmetries could be accidental and emerge only at long distances, which nonetheless cannot be predicted a priori. It is not known whether symmetry can be the outgrowth of more fundamental principles.

On the other hand, in attempting to answer the question of existence, J. A. Wheeler adopted the philosophy that every physical quantity is derived from binary bits, inventing the phrase It from bit [1]. Indeed, remarkable connections between fundamental physics and information science have been revealed in the past decades. Central to this pursuit is quantum entanglement, a basic phenomenon in quantum mechanics. Naturally one is led to wonder: can symmetry come from qubit? [2]

In this regard there is a fascinating observation in low energy Quantum Chromodynamics (QCD) recently. Ref. [3] computed the entanglement power in two-nucleon scattering and found entanglement suppression in regions of parameter space with enhanced Wigner’s spin-flavor symmetry and Schrödinger’s non-relativistic conformal invariance. This raises the intriguing possibility of understanding symmetry from quantum entanglement.

In this work we aim to purify the observation and study the correlation between symmetry and entanglement suppression in an information-theoretic setting. As a result, our conclusions are universal, quantum mechanically speaking, and not confined to QCD.

Alice and Bob Meet Cartan

We start by considering two qubits, Alice (AA) and Bob (BB), each with the basis vectors ∣0⟩|0\rangle and ∣1⟩|1\rangle. For a spin-1/21/2 system, ∣0⟩|0\rangle and ∣1⟩|1\rangle can be identified with states having the spin-component Sz=+1/2S_z=+1/2 and −1/2-1/2, respectively. It is customary to define the computational basis {∣00⟩,∣01⟩,∣10⟩,∣11⟩}\{|00\rangle, |01\rangle, |10\rangle, |11\rangle\}, where ∣ij⟩=∣i⟩A⊗∣j⟩B|ij\rangle=|i\rangle_A\otimes|j\rangle_B. There are several measures of entanglement of two-qubit systems [4] but our analysis relies only on general features of these measures, namely that they vanish if ∣Ψ⟩|\Psi\rangle is unentangled, i.e., can be written as a direct product state, ∣Ψ⟩=∣ϕ⟩⊗∣ϕ′⟩|\Psi\rangle=|\phi\rangle\otimes|\phi'\rangle for some ∣ϕ⟩,∣ϕ′⟩|\phi\rangle,|\phi'\rangle, and are maximal for maximally entangled states. Examples of maximally entangled states include the Bell basis:

∣Φ±⟩=∣00⟩±∣11⟩2,∣Ψ±⟩=∣01⟩±∣10⟩2,(1)|\Phi^{\pm}\rangle=\frac{|00\rangle\pm|11\rangle}{\sqrt{2}},\qquad|\Psi^{\pm}\rangle=\frac{|01\rangle\pm|10\rangle}{\sqrt{2}}, \tag*{(1)}

as well as the magic basis [5], defined in the same way as the Bell basis except for some phases: ∣Φ1⟩=∣Φ+⟩|\Phi_1\rangle=|\Phi^+\rangle, ∣Φ2⟩=−i∣Φ−⟩|\Phi_2\rangle=-i|\Phi^-\rangle, ∣Φ3⟩=∣Ψ+⟩|\Phi_3\rangle=|\Psi^+\rangle, ∣Φ4⟩=−i∣Ψ−⟩|\Phi_4\rangle=-i|\Psi^-\rangle. An example of an entanglement measure is the concurrence Δ(Φ)\Delta(\Phi) [5, 6], which for a normalized state ∣Ψ⟩=α∣00⟩+β∣10⟩+γ∣01⟩+δ∣11⟩|\Psi\rangle=\alpha|00\rangle+\beta|10\rangle+\gamma|01\rangle+\delta|11\rangle is defined as

Δ(Ψ)=∣⟨Ψ∣σy⊗σy∣Ψ∗⟩∣=2∣αδ−βγ∣.(2)\Delta(\Psi)=|\langle\Psi|\sigma_y\otimes\sigma_y|\Psi^*\rangle|=2|\alpha\delta-\beta\gamma|. \tag*{(2)}

The concurrence has the property that 0≤Δ(Ψ)≤10\leq\Delta(\Psi)\leq1, where Δ(Ψ)=0\Delta(\Psi)=0 if ∣Ψ⟩|\Psi\rangle is not entangled and Δ(Ψ)=1\Delta(\Psi)=1 if it is maximally entangled. Concurrence is reviewed in the Appendix and its relation to some other entanglement measures is discussed. The crucial property that all these measures share is that the entanglement is unchanged by single qubit unitary operators.

A quantum-mechanical operator acting on the two-qubit system is represented by a matrix UU in the SU(4)SU(4) group, which is parameterized by 15 generators:

{σa⊗12, 12⊗σb, σa⊗σb; a,b=x,y,z},(3)\{\sigma^{a}\otimes\mathbf{1}_{2},\ \mathbf{1}_{2}\otimes\sigma^{b},\ \sigma^{a}\otimes\sigma^{b};\ a,b=x,y,z\}, \tag*{(3)}

where σa\sigma^{a}’s are the Pauli matrices and 12\mathbf{1}_{2} is the 2×22 \times2 identity matrix. It turns out that not all UU’s generate entanglement; operators that are local in the product space,

U=UA⊗UB,UA/B∈SU(2),U=U_{A}\otimes U_{B},\quad U_{A/B}\in SU(2),

leave the entanglement unchanged, leading to the notion of equivalent classes: two operators UU and U′U^{\prime} generate the same amount of entanglement if they are related by local operators,

U∼U′,ifU=(UA⊗UB)U′(VA⊗VB),(4)U\sim U^{\prime},\quad\text{if}\quad U=(U_{A}\otimes U_{B})U^{\prime}(V_{A}\otimes V_{B}), \tag*{(4)}

for some single qubit operators UA/BU_{A/B} and VA/BV_{A/B}.

Classification of all non-local, and hence entanglement generating operators in a two-qubit system has been achieved long ago [4, 7, 8] and relies on the Cartan decomposition [9] of the Lie algebra of a symmetric coset space G/HG/H, whose generators satisfy

[h,h]=h,[h,p]=p,[p,p]=h,(5)[h,h]=h,\quad[h,p]=p,\quad[p,p]=h, \tag*{(5)}

where h={Ti}h=\{T^{i}\} and p={Xa}p=\{X^{a}\} are generators of HH and G/HG/H, respectively. The Lie algebra g\mathfrak{g} of GG can be written as g=h⊕p\mathfrak{g}=h\oplus p. Let a={Aa}a=\{A^{a}\} be the maximal abelian subalgebra of g\mathfrak{g} contained in pp: a⊂pa\subset p and [a,a]=0[a,a]=0. Then an element gg of GG can be written as

g=K1eiβaAaK2,K1,K2⊂H,(6)g=K_{1}e^{i\beta_{a}A^{a}}K_{2},\quad K_{1},K_{2}\subset H, \tag*{(6)}

for some real parameters {βa}\{\beta_{a}\}. This is generally referred to as the Cartan decomposition for symmetric spaces.

The Cartan decomposition in Eq. (6) can be understood as follows. The coset structure allows for the “polar decomposition” for any g∈Gg\in G,

g=Kˉ1L,Kˉ1∈H and L∈G/H.(7)g=\bar{K}_{1}L,\quad\bar{K}_{1}\in H\ \text{and}\ L\in G/H. \tag*{(7)}

Next we would like to show that,

p=a⊕[h,a].(8)p=a\oplus[h,a]. \tag*{(8)}

Recall that, since a⊂pa\subset p, Eq. (5) implies [h,a]⊂p[h,a]\subset p. Consider a generator Xa∈pX^{a}\in p that is orthogonal to [h,a][h,a],

0=Tr⁡(Xa[h,a])=Tr⁡(h[a,Xa]),(9)0=\operatorname{Tr}(X^{a}[h,a])=\operatorname{Tr}(h[a,X^{a}]), \tag*{(9)}

where we have used the cyclic property of the trace in the second equality. Using Eq. (5) again, [Xa,a]⊂h[X^{a},a]\subset h which implies the only way Eq. (9) can hold is if [Xa,a]=0[X^{a},a]=0. This shows XaX^{a} must be part of the maximal Cartan subalgebra a⊂pa\subset p and Eq. (8) follows.

The last step is to apply Baker-Campbell-Hausdorff formula,

eiZeiYe−iZ=ei(Y+[Z,Y]+12![Z,[Z,Y]]+13![Z,[Z,[Z,Y]]]+⋯ ),(10)e^{iZ}e^{iY}e^{-iZ}=e^{i\left(Y+[Z,Y]+\frac{1}{2!}[Z,[Z,Y]]+\frac{1}{3!}[Z,[Z,[Z,Y]]]+\cdots\right)}, \tag*{(10)}

with Y=aY=a and Z=hZ=h. Using Eqs. (5) and (8) we see [h,[h,a]]=p[h,[h,a]]=p, [h,[h,[h,a]]]=p[h,[h,[h,a]]]=p, etc. Therefore the right-hand side of Eq. (10) sits in G/HG/H, leading to

L=(K2)−1eiβaAaK2.(11)L=(K_{2})^{-1}e^{i\beta_{a}A^{a}}K_{2}. \tag*{(11)}

Plugging Eq. (11) into Eq. (7) we arrive at Eq. (6).

Minimal Entanglers

For SU(4)/(SU(2)⊗SU(2))SU(4)/(SU(2)\otimes SU(2)), with the SU(2)SU(2)’s generated by σa⊗12\sigma^{a}\otimes\mathbf{1}_{2} and 12⊗σb\mathbf{1}_{2}\otimes\sigma^{b}, a={σa⊗σa, a=x,y,z}a=\{\sigma^{a}\otimes\sigma^{a},\ a=x,y,z\}. The space of inequivalent “entanglers” is then parameterized by only 3 parameters, (βx,βy,βz)(\beta_{x},\beta_{y},\beta_{z}),

Ud=eiβaσa⊗σa=(cxcycz+isxsysz)1+[(sxsycz+icxcysz)σx⊗σxU_{d}=e^{i\beta_{a}\sigma^{a}\otimes\sigma^{a}}=(c_{x}c_{y}c_{z}+i s_{x}s_{y}s_{z})\mathbf{1} +[(s_{x}s_{y}c_{z}+i c_{x}c_{y}s_{z})\sigma^{x}\otimes\sigma^{x}
+(cyclic perm. of {x,y,z})],(12)\qquad+(cyclic\ perm.\ of\ \{x,y,z\})], \tag*{(12)}

where ca=cos⁡βac_{a}=\cos\beta_{a} and sa=sin⁡βas_{a}=\sin\beta_{a} and 1\mathbf{1} is the four-dimensional identity matrix. It is worth recalling some basic properties of UdU_{d}. First, the amount of entanglement created by UdU_{d} is periodic in βi→βi+π/2\beta_{i}\to\beta_{i}+\pi/2 [4]. To see this, consider

ei(βx+δ)σx⊗σx=(1cos⁡δ+iσx⊗σxsin⁡δ)eiβxσx⊗σx.(13)e^{i(\beta_{x}+\delta)\sigma^{x}\otimes\sigma^{x}}=(\mathbf{1}\cos\delta+i\sigma^{x}\otimes\sigma^{x}\sin\delta)e^{i\beta_{x}\sigma^{x}\otimes\sigma^{x}}. \tag*{(13)}

When δ=π/2\delta=\pi/2 the above operator is related to the operator with δ=0\delta=0 by an operator that is a product of local unitaries, UA⊗UB=iσx⊗σxU_{A}\otimes U_{B}=i\sigma^{x}\otimes\sigma^{x}, and therefore belongs to the same equivalent class. Moreover, the eigenvalues of UdU_{d} (most easily computed in the Bell basis which is the eigenbasis of UdU_{d}) are λi=eiϕi\lambda_{i}=e^{i\phi_{i}}, where ϕi\phi_{i} are

ϕ1=βx−βy+βz,ϕ2=−βx+βy+βz\phi_{1}=\beta_{x}-\beta_{y}+\beta_{z},\qquad\phi_{2}=-\beta_{x}+\beta_{y}+\beta_{z}
ϕ3=βx+βy−βz,ϕ4=−βx−βy−βz.(14)\phi_{3}=\beta_{x}+\beta_{y}-\beta_{z},\qquad\phi_{4}=-\beta_{x}-\beta_{y}-\beta_{z}. \tag*{(14)}

These are the eigenvalues of ∣Φ+⟩\lvert\Phi^{+}\rangle, ∣Φ−⟩\lvert\Phi^{-}\rangle, ∣Ψ+⟩\lvert\Psi^{+}\rangle, and ∣Ψ−⟩\lvert\Psi^{-}\rangle, respectively. Then (βx,βy)→(−βx,−βy)(\beta_{x},\beta_{y})\to(-\beta_{x},-\beta_{y}) or (βx,βy)→(βy,βx)(\beta_{x},\beta_{y})\to(\beta_{y},\beta_{x}) generate all possible permutations of the eigenvalues. Permuting the eigenvalues of UdU_{d} does not change the entanglement properties of UdU_{d}. The geometric structure underlying all UdU_{d} is a 3-torus, T3=S1×S1×S1T^{3}=S^{1}\times S^{1}\times S^{1}, and we can restrict to π/2≥βx≥βy≥βz≥0\pi/2\geq\beta_{x}\geq\beta_{y}\geq\beta_{z}\geq0, π/2≥βx+βy\pi/2\geq\beta_{x}+\beta_{y}. If βz=0\beta_{z}=0 we have the additional constraint π/4≥βx\pi/4\geq\beta_{x}. This corresponds to one Weyl chamber of G/HG/H [8, 10], where every point corresponds to an equivalent class. We will use [Ud][U_{d}] to denote the equivalent class containing UdU_{d}. Related discussions in the context of the SS-matrix in a quantum field theory can be found in Ref. [11].

To quantify the ability of an operator to produce entanglement, it is possible to define the entanglement power of an operator by averaging over the states it acts on [12]. For our purpose, it is sufficient to consider Ud∣ψ⟩U_d|\psi\rangle, where ∣ψ⟩=(a∣0⟩+b∣1⟩)⊗(c∣0⟩+d∣1⟩)|\psi\rangle=(a|0\rangle+b|1\rangle)\otimes(c|0\rangle+d|1\rangle) is a general product state in the computational basis, compute the concurrence defined in Eq. (2) and average over the wave function of the qubit on the Bloch sphere. It turns out that it is easier to compute the average of the square of the concurrence [13, 14], as explained in the Supplementary Material,

12Δ(Ud∣ψ⟩)2‾=16−118∑a<bcos⁡(4βa)cos⁡(4βb).(15)\frac{1}{2}\overline{\Delta(U_d|\psi\rangle)^2}=\frac{1}{6}-\frac{1}{18}\sum_{a<b}\cos(4\beta_a)\cos(4\beta_b). \tag*{(15)}

Since Eq. (15) is positive-definite, a vanishing entanglement power requires cos⁡(4βa)cos⁡(4βb)=1\cos(4\beta_a)\cos(4\beta_b)=1 for all aa and bb. The only solutions are βx=βy=βz=0\beta_x=\beta_y=\beta_z=0 or π/4\pi/4. So there are only two minimally entangling UdU_d whose matrix representations in the computational basis are,

βa=0:Ud=1,\beta_a=0: U_d=1,
βa=π4:Ud=eiπ4[1000001001000001]≡eiπ4×(SWAP).(16)\beta_a=\frac{\pi}{4}: U_d=e^{i\frac{\pi}{4}} \begin{bmatrix} 1&0&0&0\\ 0&0&1&0\\ 0&1&0&0\\ 0&0&0&1 \end{bmatrix} \equiv e^{i\frac{\pi}{4}}\times(\mathrm{SWAP}). \tag*{(16)}

The SWAP interchanges the states of qubit AA and qubit BB, i.e., ∣ij⟩→∣ji⟩|ij\rangle\to|ji\rangle in the computational basis and SWAP2=1\mathrm{SWAP}^2=1. The corresponding equivalence classes are [1][1] and [SWAP][\mathrm{SWAP}].

Two-body scattering

In a scattering experiment AB→ABAB\to AB, where AA and BB are fermions, the SS-matrix can be considered as an entanglement operator in the spin-space. Moreover, it is well-known that the low-energy scattering of non-identical fermions is dominated by the ss-wave channel, which we focus on here. See Refs. [15, 16] for prior discussions of entanglement in fermion scattering.

In the ss-wave channel, there is no orbital angular momentum and the SS-matrix, SS, can be decomposed into the singlet, 1S0{}^{1}S_{0}, and the triplet, 3S1{}^{3}S_{1}, channels [17], S=S0+S1S=S_0+S_1. The general form of SS is constrained by 1) unitarity: S†S=1S^\dagger S=1, and 2) rotational invariance: [Si,J⃗]=0[S_i,\vec{J}]=0, where in the ss-wave J⃗=s⃗A+s⃗B\vec{J}=\vec{s}_A+\vec{s}_B is the total spin of AA and BB. In the computational basis, s⃗A=(σ⃗/2)⊗12\vec{s}_A=(\vec{\sigma}/2)\otimes1_2 and s⃗B=12⊗(σ⃗/2)\vec{s}_B=1_2\otimes(\vec{\sigma}/2). Therefore SiS_i can only be a linear combination of the identity 11 and J2=(s⃗A+s⃗B)2J^2=(\vec{s}_A+\vec{s}_B)^2:

Si=ei2δi(ai1+bi σ⃗⋅σ⃗),σ⃗⋅σ⃗≡∑aσa⊗σa.(17)S_i=e^{i2\delta_i}(a_i1+b_i\,\vec{\sigma}\cdot\vec{\sigma}),\qquad\vec{\sigma}\cdot\vec{\sigma}\equiv\sum_a\sigma^a\otimes\sigma^a. \tag*{(17)}

where δi=δi(p)\delta_i=\delta_i(p) is a phase which could depend on the centre-of-mass momentum pp and ai,bia_i,b_i are constants. Demanding that S0S_0 and S1S_1 project onto the singlet and the triplet channels, respectively, gives a1=3b1a_1=3b_1 and a0=−b0a_0=-b_0. The overall normalization is fixed by the unitarity constraint, S†S=1S^\dagger S=1, and we arrive at

S=14(3ei2δ1+ei2δ0)1+14(ei2δ1−ei2δ0)σ⃗⋅σ⃗,(18)S=\frac{1}{4}(3e^{i2\delta_1}+e^{i2\delta_0})1+\frac{1}{4}(e^{i2\delta_1}-e^{i2\delta_0})\vec{\sigma}\cdot\vec{\sigma}, \tag*{(18)}
=12(ei2δ1+ei2δ0)1+12(ei2δ1−ei2δ0)SWAP,(19)=\frac{1}{2}(e^{i2\delta_1}+e^{i2\delta_0})1+\frac{1}{2}(e^{i2\delta_1}-e^{i2\delta_0})\mathrm{SWAP}, \tag*{(19)}

where we have used SWAP=(1+σ⃗⋅σ⃗)/2\mathrm{SWAP}=(1+\vec{\sigma}\cdot\vec{\sigma})/2. Surprisingly, rotational invariance and unitarity constrains the ss-wave SS-matrix to be a linear combination of minimal entanglers. Realizing a minimally entangling gate in the SS-matrix then requires the relations:

δ0(p)−δ1(p)=0⇒S∼[1],(20)\delta_0(p)-\delta_1(p)=0\qquad\Rightarrow\qquad S\sim[1], \tag*{(20)}
∣δ0(p)−δ1(p)∣=π/2⇒S∼[SWAP].(21)|\delta_0(p)-\delta_1(p)|=\pi/2\qquad\Rightarrow\qquad S\sim[\mathrm{SWAP}]. \tag*{(21)}

Given the momentum dependence in δi\delta_i's, the above relations are not satisfied generically, except at certain fixed momenta. Therefore, when the relations hold independently of the momentum pp, we are sitting in special regions of parameter space. Indeed, these are regions where symmetries emerge.

At sufficiently low energies, when the effective-range expansion is valid, pcot⁡δi(p)p\cot\delta_i(p) is analytic in p2p^2

pcot⁡δi(p)=−1ai+12r0(i)p2+⋯ ,(22)p\cot\delta_i(p)=-\frac{1}{a_i}+\frac{1}{2}r_0^{(i)}p^2+\cdots, \tag*{(22)}

where aia_i is the scattering length and r0(i)r_0^{(i)} is the effective range. This translates into

tan⁡δi(p)=−pai+O(p3).(23)\tan\delta_i(p)=-pa_i+\mathcal{O}(p^3). \tag*{(23)}

The limit ai→0a_i\to0 and δi→0\delta_i\to0 corresponds to a free theory in that channel, while ai→±∞a_i\to\pm\infty and δi→∓π/2\delta_i\to\mp\pi/2 arises from the “unitarity limit”, with the largest possible cross section consistent with unitarity. Then it is clear that to achieve the [SWAP] gate with ∣δ0−δ1∣=π/2|\delta_0-\delta_1|=\pi/2 for more than one pp one must tune a→±∞a\to\pm\infty in one channel and a→0a\to0 in the other channel. This corresponds to an RG fixed point of the theory, which is known to possess an emergent Schrödinger symmetry [18].

On the other hand, realizing the Identity with δ0(p)=δ1(p)\delta_0(p)=\delta_1(p) one needs to set the scattering parameters in the two channels equal. This could be enforced through an internal symmetry, which we turn to next.

Effective Hamiltonians

In this section we investigate what kind of Hamiltonians would give rise to the minimally entangling [1] gate in the SS-matrix. In a quantum field theory the phase δi\delta_i is computed through the scattering amplitude Ai\mathcal{A}_i defined by e2iδi=1+i(Mp/2π)Aie^{2i\delta_i}=1+i(Mp/2\pi)\mathcal{A}_i, where MM is the mass of the fermions (which we take to be degenerate, for simplicity). We then have

Ai=4πM1pcot⁡δi−ip.(24)\mathcal{A}_i = \frac{4\pi}{M}\frac{1}{p\cot\delta_i-ip}. \tag*{(24)}

To realize the Identity gate we need δ0=δ1\delta_0=\delta_1, which implies the amplitudes in both channels are equal, suggesting the existence of an internal symmetry.

In what follows we will consider an effective Hamiltonian for two species of qubits, labelled by the species index: I,J=1,2I,J=1,2. Then the notation ∣Ii;Jj⟩\lvert I i;J j\rangle represents the situation where Alice possesses a qubit ∣i⟩\lvert i\rangle in species II and Bob has a qubit ∣j⟩\lvert j\rangle in species JJ. The addition of the species quantum number opens the possibility of entanglement through the species quantum numbers.

An entangler that takes an input ∣Ii;Jj⟩\lvert I i;J j\rangle to an output ∣I′i′;J′j′⟩\lvert I' i';J' j'\rangle can arise from the generic Hamiltonian,

aI′i′†aIiaJ′j′†aJj,(25)a_{I'i'}^\dagger a_{Ii}a_{J'j'}^\dagger a_{Jj}, \tag*{(25)}

where aIi†a_{Ii}^\dagger (aIia_{Ii}) creates (annihilates) ∣Ii⟩\lvert Ii\rangle and all other quantum numbers (such as the momentum) are suppressed. The most general effective Hamiltonian is

Hint=∑Gii′jj′KII′JJ′aI′i′†aIiaJ′j′†aJj+h.c.,(26)H_{\mathrm{int}}=\sum G_{ii'jj'}K_{II'JJ'}a_{I'i'}^\dagger a_{Ii}a_{J'j'}^\dagger a_{Jj}+\mathrm{h.c.}, \tag*{(26)}

where Gii′jj′G_{ii'jj'} and KII′JJ′K_{II'JJ'} are coupling constants in the spin- and species-space, respectively.

From Eq. (16) we see that, in order for HintH_{\mathrm{int}} to generate [1], it must be diagonal in the spin-space, Gii′jj′=Gijδii′δjj′G_{ii'jj'}=G_{ij}\delta_{ii'}\delta_{jj'}. Rotational invariance further implies GijG_{ij} must be independent of (ij)(ij),

G=gs(1111),(27)\mathcal{G}=g_s\begin{pmatrix}1&1\\1&1\end{pmatrix}, \tag*{(27)}

where gsg_s is a constant. It turns out to achieve entanglement suppression when there is more than one species of qubit, HintH_{\mathrm{int}} must conserve the “species number” for each individual species [3]. To see this, consider the following species-changing one-body operators,

Θ=aI↑†aJ↑+h.c..(28)\Theta=a_{I\uparrow}^{\dagger}a_{J\uparrow}+\mathrm{h.c.}. \tag*{(28)}

When Θ\Theta acts on the state ∣I↑;J↑⟩\lvert I\uparrow;J\uparrow\rangle,

Θ∣I↑;J↑⟩=∣I↑;I↑⟩+∣J↑;J↑⟩,(29)\Theta\lvert I\uparrow;J\uparrow\rangle=\lvert I\uparrow;I\uparrow\rangle+\lvert J\uparrow;J\uparrow\rangle, \tag*{(29)}

which will result in an entangled output. Therefore, entanglement suppression also requires KII′,JJ′=KIJδII′δJJ′K_{II',JJ'}=K_{IJ}\delta_{II'}\delta_{JJ'}. If we use the notation NI=(aI↑,aI↓)TN_I=(a_{I\uparrow},a_{I\downarrow})^{T}, the Hamiltonian can now be written as

Hint=gs∑I,JKIJ(NI†NI)(NJ†NJ)+h.c.,(30)H_{\mathrm{int}}=g_s\sum_{I,J}K_{IJ}\left(N_I^\dagger N_I\right)\left(N_J^\dagger N_J\right)+\mathrm{h.c.}, \tag*{(30)}

which is invariant under an SU(2)1×SU(2)2SU(2)_1\times SU(2)_2 global symmetry,

NI→UINI,UI∈U(2)I,I=1,2.(31)N_I\to U_IN_I,\qquad U_I\in U(2)_I,\quad I=1,2. \tag*{(31)}

The rotational invariance is the diagonal subgroup of SU(2)1×SU(2)2SU(2)_1\times SU(2)_2, which acts on each species identically.

In some cases it is possible to further assume the existence of a “species symmetry” such that, much like the rotational invariance in Eq. (27), the coupling constant KIJK_{IJ} is universal and independent of (IJ)(IJ),

K=gf(1111).(32)\mathcal{K}=g_f\begin{pmatrix}1&1\\1&1\end{pmatrix}. \tag*{(32)}

In the context of two-nucleon scattering the universality is achieved through the SU(2)SU(2) isospin symmetry transforming the neutron and the proton into each other. In this case the Hamiltonian becomes

Hint=gsgf(N†N)2,(33)H_{\mathrm{int}}=g_sg_f\left(\mathcal{N}^{\dagger}\mathcal{N}\right)^2, \tag*{(33)}

where N=(N1,N2)T=(a1↑,a1↓,a2↑,a2↓)T\mathcal{N}=(N_1,N_2)^T=(a_{1\uparrow},a_{1\downarrow},a_{2\uparrow},a_{2\downarrow})^T. We see the SU(2)1×SU(2)2SU(2)_1\times SU(2)_2 global symmetry is enlarged to an SU(4)SU(4) symmetry. This is Wigner’s SU(4)SU(4) spin-flavor symmetry in effective interactions of nucleons [19].

The above argument generalizes immediately to the case when there are NqN_q species of qubit: I,J=1,2,…,NqI,J=1,2,\ldots,N_q. In fact, Eq. (30) remains valid in the general case and the global symmetry is NqN_q-copy of SU(2)SU(2): SU(2)1×⋯SU(2)NqSU(2)_1\times\cdots SU(2)_{N_q}. Moreover, when there is a species-universal coupling constant (due to some underlying species symmetry), the Hamiltonian implementing the [1] gate must be diagonal in both the spin- and species-space, with N\mathcal{N} in Eq. (33) now defined by N=(a1↑,a1↓,…,aNq↑,aNq↓)T\mathcal{N}=(a_{1\uparrow},a_{1\downarrow},\ldots,a_{N_q\uparrow},a_{N_q\downarrow})^T. The resulting global symmetry is enlarged to SU(2Nq)SU(2N_q).

In low energy QCD interactions of the octet baryons indeed exhibit an enhanced SU(16)SU(16) spin-flavor symmetry [3]. Now we see this is a direct consequence of QCD realizing the [1] gate. The example also serves to demonstrate that, to arrive at a species-universal coupling constant among the NqN_q qubit, it is not necessary to impose an SU(Nq)SU(N_q) species symmetry. In fact, it suffices if the NqN_q species of qubit furnish an irreducible linear representation of an underlying group GqG_q which may be smaller than SU(Nq)SU(N_q). For the octet baryon this is achieved through the SU(3)SU(3) flavor symmetry in QCD.

It is interesting to entertain the possibility of realizing the [SWAP], which requires the SS-matrix in one spin channel saturating the unitarity limit (δ=π/2\delta=\pi/2) and the scattering being absent in the other channel (δ=0\delta=0). Both SS matrices are invariant under the Schrödinger invariance [18], which is not apparent at the level of the Lagrangian, but emerges after resumming certain contact interactions to all orders and performing a fine tuning of the theory parameters. For octet baryons the interactions in question are the two double trace operators in the chiral Lagrangian [20]. The resulting theory is not SU(16)SU(16) symmetric and does not appear to be realized in low energy QCD.

Conclusion

In this work we initiated an information-theoretic study on the origin of symmetry. In particular, we studied the correlation between the emergence of global symmetries and entanglement suppression in the two-body scattering of fermions in the ss-wave channel. After showing there are only two minimally entangling gates in a two-qubit system: the Identity and the SWAP gates, which turned out to be rotationally invariant, we argued the SS-matrix is constrained by unitarity and rotational invariance to be a linear combination of the Identity and the SWAP. Implementing a minimally entangling SS-matrix requires certain conditions on the phases in the 1S0{}^{1}S_{0} and 3S1{}^{3}S_{1} channels. The SWAP gate can be realized when one of the channels flows to a free fixed point in the IR and the other a unitarity fixed point with the Schrödinger symmetry. On the other hand, an SS-matrix implementing the Identity gate implies [SU(2)]Nq[SU(2)]^{N_q} global symmetry in the effective Hamiltonian, where NqN_q is the number of species of qubits. When there is a universal coupling among the species, the symmetry is enlarged to SU(2Nq)SU(2N_q).

In nature the Identity gate is realized in low energy QCD. One wonders whether the SWAP gate arises naturally in other physical systems [21]. Furthermore, while we focused on symmetries realized in the Wigner-Weyl mode [22], it would be interesting to study whether spontaneously broken symmetries, the Nambu-Goldstone mode, can also be realized in an information-theoretic setting.

Acknowledgement – I. L. is supported in part by the U.S. Department of Energy under contracts No. DE-AC02-06CH11357 at Argonne and No. DE-SC0010143 at Northwestern. He also acknowledges the hospitalities at the National Center for Theoretical Sciences at National Taiwan University and National Tsing Hua University in Taiwan, where part of this work was performed. T. M. acknowledges enlightening conversations with I. Marvian and Q. Liu. T.M. is supported by the U.S. Department of Energy, Office of Science, Office of Nuclear Physics under grant Contract Numbers DE-FG02-05ER41368 and DE-FG02-05ER41367.

Entanglement Measures

This appendix gives a brief discussion of entanglement measures, which we argue can be motivated by demanding that they be invariant under unitary transformations on single qubits. The concurrence [6] turns out to be the simplest such invariant. We then discuss its relation to other entanglement measures based on reduced density matrices, such as the von Neumann entropy and the 2-entropy. Entanglement measures for operators are defined by applying them to unentangled direct product states and averaging over all such states. This is used to prove that all minimal entanglers are equivalent to 1 or SWAP up to local unitary operations.

For a two-qubit system the most general state is

∣Ψ⟩=α∣00⟩+β∣01⟩+γ∣10⟩+δ∣11⟩.(34)|\Psi\rangle= \alpha|00\rangle+ \beta|01\rangle+ \gamma|10\rangle+ \delta|11\rangle. \tag*{(34)}

Any entanglement measure should vanish on direct product states and attain a maximum on maximally entangled states. It is easy to show that Δ(Ψ)=2∣αδ−βγ∣\Delta(\Psi) = 2|\alpha\delta- \beta\gamma| defined in Eq. (2) vanishes on all direct product states:

∣ψ⟩=(a∣0⟩+b∣1⟩)⊗(c∣0⟩+d∣1⟩).(35)|\psi\rangle= (a|0\rangle+ b|1\rangle) \otimes(c|0\rangle+ d|1\rangle) . \tag*{(35)}

In fact we will show that αδ−βγ\alpha\delta- \beta\gamma is the only second order polynomial in α,β,γ,\alpha,\beta,\gamma, and δ\delta that has this property. It is also clear that ∣αδ−βγ∣≤∣α∣∣δ∣+∣β∣∣γ∣≤(∣α∣2+∣β∣2+∣γ∣2+∣δ∣2)/2=1/2|\alpha\delta-\beta\gamma| \leq|\alpha||\delta| + |\beta||\gamma| \leq(|\alpha|^2+|\beta|^2+|\gamma|^2+|\delta|^2)/2 = 1/2, so that 0≤Δ(Ψ)≤10 \leq\Delta(\Psi) \leq1. Finally Δ(Ψ)=1\Delta(\Psi)=1 for Bell states, so Δ(Ψ)\Delta(\Psi) has the desired properties.

Here we want to emphasize another way of motivating Δ(Ψ)\Delta(\Psi), which could easily be generalized to mm-level qubits. Under a unitary transformation UAU_A on Alice the components of the wave function transform as:

\begin{pmatrix} \alpha\\ \gamma\end{pmatrix} \to U_A \begin{pmatrix} \alpha\\ \gamma\end{pmatrix}, \qquad \begin{pmatrix} \beta\\ \delta\end{pmatrix} \to U_A \begin{pmatrix} \beta\\ \delta\end{pmatrix}, \tag*{(36)}\end{pmatrix}

while under a unitary transformation UBU_B on Bob the components transform as

\begin{pmatrix} \alpha\\ \beta\end{pmatrix} \to U_B \begin{pmatrix} \alpha\\ \beta\end{pmatrix}, \qquad \begin{pmatrix} \gamma\\ \delta\end{pmatrix} \to U_B \begin{pmatrix} \gamma\\ \delta\end{pmatrix}. \tag*{(37)}\end{pmatrix}

If we assemble the components of the wave function into a matrix

M≡(αβγδ),(38)M \equiv\begin{pmatrix} \alpha& \beta\\ \gamma& \delta\end{pmatrix}, \tag*{(38)}

these transformations can be written as

M→UAMUBT.(39)M \to U_A M U_B^{T}. \tag*{(39)}

An important ingredient underlying this work is that local unitary transformations do not affect entanglement and this led us to the Cartan decomposition of a general SU(4)SU(4) element acting on two qubits. In the present context we see that entanglement measures must be invariants under the transformation in Eq. (39). There are only two invariants that are second order:

Det⁡[M]=αδ−βγ\operatorname{Det}[M] = \alpha\delta- \beta\gamma
Tr⁡[M†M]=∣α∣2+∣β∣2+∣γ∣2+∣δ∣2=1.(40)\operatorname{Tr}[M^\dagger M] = |\alpha|^2 + |\beta|^2 + |\gamma|^2 + |\delta|^2 = 1 . \tag*{(40)}

Given this, it is not surprising that other entanglement measures for two qubits can be expressed in terms of Δ[Ψ]\Delta[\Psi]. For example, the reduced density matrices for qubits AA and BB are:

ρA=MM†,ρB=M†M,(41)\rho_A = MM^\dagger, \qquad\rho_B = M^\dagger M, \tag*{(41)}

whose eigenvalues are

λ±=12(1±1−Δ2(Ψ)).(42)\lambda_{\pm} = \frac{1}{2}\left(1 \pm\sqrt{1-\Delta^{2}(\Psi)}\right). \tag*{(42)}

Any entanglement measure derived from reduced density matrices can therefore be expressed in terms of the concurrence. For example the von Neumann entropy is

EE(∣Ψ⟩)=−Tr⁡[ρAln⁡ρA]=−λ+ln⁡λ+−λ−ln⁡λ−,(43)\mathcal{E}_{E}(|\Psi\rangle)=-\operatorname{Tr}[\rho_{A}\ln\rho_{A}]=-\lambda_{+}\ln\lambda_{+}-\lambda_{-}\ln\lambda_{-}, \tag*{(43)}

while the 2-entropy is

ER(∣Ψ⟩)=1−Tr⁡[ρA2]=12Δ2(Ψ)=2∣αδ−βγ∣2.(44)\mathcal{E}_{R}(|\Psi\rangle)=1-\operatorname{Tr}[\rho_{A}^{2}]=\frac{1}{2}\Delta^{2}(\Psi)=2|\alpha\delta-\beta\gamma|^{2}. \tag*{(44)}

These share the property of Δ(Ψ)\Delta(\Psi) that they vanish when evaluated on an unentangled state. For maximally entangled states, EE=ln⁡2\mathcal{E}_{E}=\ln2 and ER=1/2\mathcal{E}_{R}=1/2.

So far we have discussed entanglement measures for states. For operators the entanglement generated is dependent on the state they act on. For example, CNOT acting on the computational basis {∣00⟩,∣01⟩,∣10⟩,∣11⟩}\{|00\rangle,|01\rangle,|10\rangle,|11\rangle\} yields {∣00⟩,∣01⟩,∣11⟩,∣10⟩}\{|00\rangle,|01\rangle,|11\rangle,|10\rangle\} and generates no entanglement. But acting on the direct product state (∣0⟩+∣1⟩)⊗∣0⟩/2(|0\rangle+|1\rangle)\otimes|0\rangle/\sqrt{2} the CNOT gate creates the maximally entangled state ∣Φ+⟩|\Phi^{+}\rangle. Forming a sensible entanglement measure for operators requires some sort of averaging over initial states. One possible approach is to average over all direct product states [12, 13], which we will do below.

We parametrize the wave function of the qubits using Bloch sphere coordinates,

a=cos⁡θA2,b=sin⁡θA2eiϕA,a=\cos\frac{\theta_{A}}{2},\quad b=\sin\frac{\theta_{A}}{2}e^{i\phi_{A}},
c=cos⁡θB2,d=sin⁡θB2eiϕB.(45)c=\cos\frac{\theta_{B}}{2},\quad d=\sin\frac{\theta_{B}}{2}e^{i\phi_{B}}. \tag*{(45)}

It is simple to compute the concurrence for arbitrary UdU_{d}. Expanding the pure state ∣ψ⟩|\psi\rangle in Eq. (35) in terms of the Bell states, which are eigenstates of UdU_{d}, it is straightforward to show:

∣αδ−βγ∣=14∣∑i=14viλi2∣,(46)|\alpha\delta-\beta\gamma|=\frac{1}{4}\left|\sum_{i=1}^{4}v_{i}\lambda_{i}^{2}\right|, \tag*{(46)}

where λi=eiϕi\lambda_{i}=e^{i\phi_{i}} is defined through the ϕi\phi_{i} in Eq. (14) and

v1=(ac+bd)2,v2=−(ac−bd)2,v_{1}=(ac+bd)^{2},\qquad v_{2}=-(ac-bd)^{2},
v3=−(ad+bc)2,v4=(ad−bc)2.(47)v_{3}=-(ad+bc)^{2},\qquad v_{4}=(ad-bc)^{2}. \tag*{(47)}

For averaging over direct product initial states it is more convenient to consider the entanglement power, which is the 2-entropy in Eq. (44) averaged over all direct product states,

ER(Ud∣ψ⟩)‾=2∣αδ−βγ∣2=18∣∑iviλi2∣2=18∑i∣vi∣2+18∑i≠jvivj∗‾λi2λj−2,(48)\begin{aligned} \overline{\mathcal{E}_{R}(U_{d}|\psi\rangle)}=2|\alpha\delta-\beta\gamma|^{2} \\ &=\frac{1}{8}\left|\sum_{i}v_{i}\lambda_{i}^{2}\right|^{2} \\ &=\frac{1}{8}\sum_{i}|v_{i}|^{2}+\frac{1}{8}\sum_{i\ne j}\overline{v_{i}v_{j}^{*}}\lambda_{i}^{2}\lambda_{j}^{-2}, \tag*{(48)} \end{aligned}

where the last line follows from λi∗=λi−1\lambda_{i}^{*}=\lambda_{i}^{-1}. The overline denotes averaging over the solid angles ΩA,B\Omega_{A,B}. It is simple to evaluate the integral

vivj∗‾=∫dΩA4π∫dΩB4πvivj∗=−19+49δij,(49)\overline{v_{i}v_{j}^{*}}=\int\frac{d\Omega_{A}}{4\pi}\int\frac{d\Omega_{B}}{4\pi}v_{i}v_{j}^{*}=-\frac{1}{9}+\frac{4}{9}\delta_{ij}, \tag*{(49)}

which yields [13, 14]

ER(Ud∣ψ⟩)‾=16−118∑a<bcos⁡(4βa)cos⁡(4βb).(50)\overline{\mathcal{E}_{R}(U_{d}|\psi\rangle)} =\frac{1}{6}-\frac{1}{18}\sum_{a<b}\cos(4\beta_{a})\cos(4\beta_{b}). \tag*{(50)}

The entanglement power is manifestly invariant under the transformations βa→βa+π/2\beta_{a}\to\beta_{a}+\pi/2, (βa,βb)→(βb,βa)(\beta_{a},\beta_{b})\to(\beta_{b},\beta_{a}) and (βa,βb)→(−βa,−βb)(\beta_{a},\beta_{b})\to(-\beta_{a},-\beta_{b}) discussed in the main text. To have minimal entanglement we must have all cos⁡(4βa)cos⁡(4βb)\cos(4\beta_{a})\cos(4\beta_{b}) in Eq. (50) equal to 1. This requires βa=0\beta_{a}=0 or π/4\pi/4 for all aa. The solution βa=0\beta_{a}=0 yields the identity gate, 1, and βa=π/4\beta_{a}=\pi/4 yields the SWAP gate times eiπ/4e^{i\pi/4}. Thus we see that all minimal entanglers are either equivalent to 1 or SWAP up to local unitary operations.

For βx=βy=βz≡β\beta_{x}=\beta_{y}=\beta_{z}\equiv\beta, the entanglement power simplifies to

ER(Ud∣ψ⟩)‾=sin⁡24β6.(51)\overline{\mathcal{E}_{R}(U_{d}|\psi\rangle)}=\frac{\sin^{2}4\beta}{6}. \tag*{(51)}

In this case the eigenvalues of UdU_{d} are eiβe^{i\beta} and e−3iβe^{-3i\beta}, while the eigenvalues of SS are e2iδ1e^{2i\delta_{1}} and e2iδ0e^{2i\delta_{0}}. Identifying 4β4\beta with 2(δ1−δ0)2(\delta_{1}-\delta_{0}) we recover the entanglement power derived in Ref. [3].

For βx=βy=βz≡β\beta_{x}=\beta_{y}=\beta_{z}\equiv\beta, the entanglement power sim-

References

  1. [1]J. A. Wheeler, in Proceedings III International Symposium on Foundations of Quantum Mechanics (1989) pp. 354–358.
  2. [2]In this work we will focus on global symmetries realized in the Wigner-Weyl mode, where the energy spectra of the system furnish linear representations of the underlying symmetry.
  3. [3]S. R. Beane, D. B. Kaplan, N. Klco, and M. J. Savage, Phys. Rev. Lett. 122, 102001 (2019), arXiv:1812.03138 [nucl-th].DOI
  4. [4]B. Kraus and J. I. Cirac, Phys. Rev. A 63, 062309 (2001).
  5. [5]S. Hill and W. K. Wootters, Phys. Rev. Lett. 78, 5022 (1997).
  6. [6]W. K. Wootters, Phys. Rev. Lett. 80, 2245 (1998).
  7. [7]N. Khaneja and S. Glaser, arXiv e-prints, quant-ph/0010100 (2000), arXiv:quant-ph/0010100 [quant-ph].DOI
  8. [8]J. Zhang, J. Vala, S. Sastry, and K. B. Whaley, Phys. Rev. A 67, 042313 (2003).
  9. [9]S. Helgason, Differential Geometry, Lie Groups, and Symmetric Spaces, Crm Proceedings & Lecture Notes (American Mathematical Society, 2001).
  10. [10]R. R. Tucci, arXiv:quant-ph/0507171 [quant-ph].DOI
  11. [11]S. R. Beane and R. C. Farrell, arXiv:2011.01278 [hep-th].DOI
  12. [12]P. Zanardi, Phys. Rev. A 63, 040304 (2001).
  13. [13]A. D. Ballard and Y.-S. Wu, in Cross disciplinary advances in quantum computing (American Mathematical Society, Oxford, 2011) Chap. Cartan Decomposition and Entangling Power of Braiding Quantum Gates.
  14. [14]A. T. Rezakhani, Phys. Rev. A 70, 052313 (2004).
  15. [15]N. L. HARSHMAN, International Journal of Modern Physics A 21, 1371 (2006).
  16. [16]N. L. Harshman, arXiv e-prints , quant-ph/0606011 (2006), arXiv:quant-ph/0606011 [quant-ph].arxiv.org/abs/quant-ph/0606011
  17. [17]In the 3S1 channel there is a small mixing with 3D1 due to the spin-orbital coupling, which we ignore.
  18. [18]T. Mehen, I. W. Stewart, and M. B. Wise, Phys. Lett. B 474, 145 (2000), arXiv:hep-th/9910025.DOI
  19. [19]T. Mehen, I. W. Stewart, and M. B. Wise, Phys. Rev. Lett. 83, 931 (1999), arXiv:hep-ph/9902370.DOI
  20. [20]Q. Liu, “Entanglement in low energy qcd,” Undergraduate thesis, Duke University, unpublished.
  21. [21]An interesting possibility is the scattering of Kerr black holes, which also suppresses entanglement power [23].
  22. [22]D. ... A study on maximal entanglement and local symmetries.

Paper details

vlnc:W_wyvppexb
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents