YITP-19-45

Chaos and Complexity in Quantum Mechanics

Tibra Ali(a) 1 , Arpan Bhattacharyya(b) 2 , S. Shajidul Haque(c)3 ,

Eugene H. Kim(c)4 , Nathan Moynihan(d)5 , Jeff Murugan (d)6

(a)
Perimeter Institute,
31 Caroline Street North
arXiv:1905.13534v1 [hep-th] 31 May 2019

Waterloo, Ontario, Canada, N2L 2Y5

(b) Center for Gravitational Physics,

Yukawa Institute for Theoretical Physics (YITP), Kyoto University,

Kitashirakawa Oiwakecho, Sakyo-ku, Kyoto 606-8502, Japan
(c) Department of Physics, University of Windsor,

401 Sunset Avenue

Windsor, Ontario, Canada, N9B 3P4
(d) The Laboratory for Quantum Gravity & Strings,

Department of Mathematics & Applied Mathematics,

University of Cape Town,
Private Bag, Rondebosch, 7701, South Africa

Abstract

We propose a new diagnostic for quantum chaos. We show that time evolution of complexity for
a particular type of target state can provide equivalent information about the classical Lyapunov
exponent and scrambling time as out-of-time-order correlators. Moreover, for systems that can be
switched from a regular to unstable (chaotic) regime by a tuning of the coupling constant of the
interaction Hamiltonian, we find that the complexity defines a new time scale. We interpret this
time scale as recording when the system makes the transition from regular to chaotic behaviour.

1
[email protected]
2
[email protected]
3
[email protected]
4
[email protected]
5
[email protected]
6
[email protected]
1 Introduction
Quantum chaos is intrinsically difficult to characterize. Consequently, a precise definition of quan-
tum chaos in many-body systems remains elusive and our understanding of the dynamics of quan-
tum chaotic systems is still inadequate. This lack of understanding is at the heart of a number of
open questions in theoretical physics such as thermalization and transport in quantum many-body
systems, and black hole information loss. It has also precipitated the renewed interest in quantum
chaos from various branches of physics from condensed matter physics to quantum gravity [1].

Chaotic classical systems on the other hand are characterised by their sensitive dependence on
initial conditions — two copies of such a system, prepared in nearly identical initial states (namely,
two distinct points in phase space, separated by a very small distance), will evolve over time into
widely separated configurations. More precisely, the distance between the two points in phase space
grows as exp(λL t), where λL is the system’s largest Lyapunov exponent [2]. This does not happen
in quantum mechanics — two nearly identical states, i.e. states with a large initial overlap, remain
nearly identical for all time (as their overlap is constant under unitary evolution). It has been
argued [3, 4] that a quantum analog of “sensitive dependence on initial conditions” is to consider
evolving identical states with slightly different Hamiltonians, Ĥ and Ĥ + δ Ĥ. If Ĥ is the quanti-
zation of a (classically) chaotic Hamiltonian, the states will evolve into two different states whose
inner product decays exponentially in time.

Traditionally, chaos in quantum systems has been identified by comparison with results from
random matrix theory (RMT)[5]. Recently however, other diagnostics have been proposed to probe
chaotic quantum systems [6, 7, 8]. One such diagnostic is out-of-time-order correlators (OTOCs)
[9, 10] from which both the (classical) Lyapunov exponent as well as the scrambling time [11, 12, 13]
may be extracted. However, recent work in mass-deformed SYK models [14] have revealed some
tension between the OTOC and RMT diagnostics that arise, in part, through the nature of the
probes. The OTOC captures early-time quantum mechanical features of the system while RMT
diagnostics typically capture late-time statistical features. Evidently, a deeper understanding of
probes of quantum chaos is required. In this light, it is interesting therefore to ask whether one
can characterize chaos in quantum systems using quantum information-theoretic measures7 . In this
work, we propose a new diagnostic/probe of quantum chaos using the notion of circuit complexity
[15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27], adopting Nielsen’s geometric approach [28, 29, 30].
Specifically, we study the sensitivity of the geodesics (and, in particularly, the geodesic distance)
in the space of unitaries to evolution with slightly different Hamiltonians, Ĥ and Ĥ + δ Ĥ. We
demonstrate how this enables one to probe/characterize chaotic quantum systems, giving informa-
tion beyond what is contained in the OTOC.

To establish out testing method, we consider a simple, exactly solvable system — the inverted
oscillator, described by the Hamiltonian H = p2 /2 − ω 2 x2 /2 [38] — which captures the exponential
sensitivity to initial conditions exhibited by chaotic systems [37]. Classically, the inverted oscillator
has an unstable fixed point at (x = 0, p = 0); a particle accelerates exponentially away from the fixed
7
Some progress in this direction was made in [31, 32, 33, 34, 35, 36].

1
point when perturbed. Though the phase-space volume of the inverted oscillator is unbounded,
our results are relevant to systems with a bounded phase space in that such a system would be
described by an inverted oscillator up to a certain time. The two systems produce the same results
over the time of interest (but would not be analytically solvable beyond that time). In what follows,
we include the analysis for a regular oscillator as a reference for what arises in a non-chaotic system
and explore also a many-body system (quantum field theory) where the inverted oscillator appears.
It is worth noting the inverted quantum oscillator is not just a toy model; it has been realized exper-
imentally [39] and has even played a role in mathematics, in attacking the Riemann hypothesis [40].

The rest of the paper is organized as follows. In Section 2, we present the model and states
considered in this work. In Section 3, we review the ideas behind circuit complexity, and compute
the circuit complexity for our system. Sections 4 and 5 demonstrate how quantum chaos can
be detected and quantified using circuit complexity while Section 6 discusses the OTOC and its
relation to the results obtained from the circuit complexity. In Section 7, we discuss a many-body
system (quantum field theory) where the inverted oscillator arises. Finally, we summarize and
present concluding remarks in Section 8.

2 The Model
We are interested in comparing the complexity of a regular system with that of an unstable/chaotic
system. To that end, we consider the Hamiltonian
1 Ω2 2
H = p2 + x where Ω2 = m2 − λ . (1)
2 2
For λ < m2 , equation (1) describes a simple harmonic oscillator; for λ > m2 , we have an inverted
oscillator. The λ = m2 case, of course describes a free particle. Our inverted oscillator model
can be understood as a short-time approximation for unstable/chaotic systems. In particular, this
model captures the exponential sensitivity to initial conditions exhibited by chaotic systems. Let’s
start with the following state at t = 0,
ωr x2
 
ψ(x, t = 0) = N (t = 0) exp − , (2)
2
where
ωr = m. (3)
Evolving this state in time by the Hamiltonian (1) produces [41]
ω(t) x2
 
ψ(x, t) = N (t) exp − , (4)
2
where N (t) is the normalization factor and
 
Ω − i ωr cot(Ω t)
ω(t) = Ω . (5)
ωr − i Ω cot(Ω t)
We will be computing the complexity for this kind of time evolved state (4) with respect to (2) and

ω(t = 0) = ωr . (6)

2
3 Complexity from the Covariance Matrix
We will start this section with a quick review of circuit complexity and then conclude with a
computation of the circuit complexity for a single oscillator. For circuit complexity we will use the
covariance matrix method. Note that a similar analysis can be done for circuit complexity from
the full wave function.

3.1 Review of Circuit Complexity

Here we will briefly sketch the outline of the computation of circuit complexity. Details of this can
be found in [15, 17]. We will highlight only the key formulae and interested readers are referred
to [15, 17] and citations thereof. The problem is simple enough to state; given a set of elementary
gates and a reference state, we want to build the most efficient circuit that starts at the reference
state and terminates at a specified target state. Formally,

|ψτ =1 i = Ũ (τ = 1)|ψτ =0 i, (7)

where τ
←
−
Z
Ũ (τ ) = P exp(i dτ H(τ )), (8)
0
is the unitary operator representing the quantum circuit, which takes the reference state |ψτ =0 i to
the target state |ψτ =1 i. τ parametrizes a path in the space of the unitaries and given a particular
basis (elementary gates) MI ,
H(τ ) = Y I (τ )MI .
In this context, the coefficients {Y I (τ )} are referred to as ‘control functions’. The path ordering in
(8) is necessary as all the MI ’s do not necessarily commute with each other.

Now, since the states under consideration (2) and (4) are Gaussian, they can be equivalently
described by a Covariance matrix as follows

Gab = hψ(x, t)|ξ a ξ b + ξ b ξ a |ψ(x, t)i, (9)

where ξ = {x, p}. This covariance matrix is typically a real symmetric matrix with unit determinant.
We will always transform the reference covariance matrix such that [20, 22]

G̃τ =0 = S · Gτ =0 · S T (10)

with G̃τ =0 an identity matrix and S a real symmetric matrix whose transpose is denoted S T .
Similarly, the reference state will transform as

G̃τ =1 = S · Gτ =1 · S T . (11)

The unitary Ũ (τ ) acts on this transformed covariance matrix as,

G̃τ =1 = Ũ (τ ) · G̃τ =0 · Ũ −1 (τ ). (12)

3
Next we define suitable cost function F(Ũ , Ũ˙ ) and define [15, 28, 29, 30]
Z 1
C(Ũ ) = F(Ũ , Ũ˙ ) dτ . (13)
0

Minimizing this cost functional gives us the optimal set of Y I (τ ), which in turn give us the most ef-
ficient circuit by minimizing the circuit depth. There are various possible choices for these functions
F(Ũ , Ũ˙ ). For further details, we refer the reader to the extensive literature in [15, 16, 17, 28, 29, 30].
In this paper, we will choose sX
F2 (U, Y ) = (Y I )2 . (14)
I

For this choice, one can easily see that, after minimization the C(Ũ ) defined in (13) corresponds to
the geodesic distance on the manifold of unitaries. Note also that we can reproduce our analysis
done in the following sections with other choices of cost functional. We will, however, leave this for
future work.

3.2 Circuit Complexity for a Single Oscillator

For our case, the covariance matrix corresponding to target state (4) will take the form,
Im(ω(t))
1
!
Re(ω(t)) − Re(ω(t))
Gτ =1 = |ω(t)|2 , (15)
− Im(ω(t))
Re(ω(t)) Re(ω(t))

where ω(t) is defined in (5). For the reference state (2) it will take the following form,
!
1
τ =0 ω 0
G = r . (16)
0 ωr

Next we change the basis as follows

G̃τ =1 = S · Gτ =1 · S T , G̃τ =0 = S · Gτ =0 · S T , (17)

with √ !
ωr 0
S= , (18)
0 √1
ωr

such that G̃τ =0 = I is an identity matrix. For the case under study, the reference frequency ωr is
real. We will choose the following three generators,
i i i
M11 → (x p + p x), M22 → x2 , M33 → p2 . (19)
2 2 2
These will serve as our elementary gates and satisfy the SL(2, R) algebra.

[M11 , M22 ] = 2 M22 , [M11 , M33 ] = −2 M33 , [M22 , M33 ] = M11 . (20)

Next, if we parameterize the Ũ (τ ) as,

!
cos(µ(τ )) cosh(ρ(τ )) − sin(θ(τ )) sinh(ρ(τ )) − sin(µ(τ )) cosh(ρ(τ )) + cos(θ(τ )) sinh(ρ(τ ))
Ũ (τ ) = .
sin(µ(τ )) cosh(ρ(τ )) + cos(θ(τ )) sinh(ρ(τ )) cos(µ(τ )) cosh(ρ(τ )) + sin(θ(τ )) sinh(ρ(τ ))
(21)

4
and set the boundary conditions as,

G̃τ =1 = Ũ (τ = 1) · G̃τ =0 · Ũ −1 (τ = 1), G̃τ =0 = Ũ (τ = 0) · G̃τ =0 · Ũ −1 (τ = 0) , (22)

we find that [22],

ωr2 + |ω(t)|2 ωr2 − |ω(t)|2

 
{cosh(2ρ(1)), tan(θ(1) + µ(1))} = , , {ρ(0), θ(0) + µ(0)} = {0, c}.
2 ωr Re(ω(t)) 2 ωr Im(ω(t))
(23)
Here c is an arbitrary constant. For simplicity we choose
ωr2 − |ω(t)|2
 
µ(τ = 1) = µ(τ = 0) = 0, θ(τ = 0) = θ(τ = 1) = c = tan−1 .
2 ωr Im(ω(t))
From (8) we have,  
Y I = Tr ∂τ Ũ (τ ) · Ũ (τ )−1 · (M I )T , (24)

where Tr M I .(M J )T = δ IJ . Using this we can define the metric

ds2 = GIJ dY I dY ∗J , (25)

where the GIJ = 21 δIJ is known as a penalty factor. Given the form of U (s) in (21) we will have,

ds2 = dρ2 + cosh(2ρ) cosh2 ρ dµ2 + cosh(2ρ) sinh2 ρ dθ2 − sinh(2ρ)2 dµ dθ, (26)

and the complexity functional defined in (13) will take the form,
Z 1 q
C(Ũ ) = dτ gij ẋi ẋj . (27)
0

The simplest solution for the geodesic is again a straight line on this geometry [15, 22]. Evaluating
(27) we simply get
 2
ωr + |ω(t)|2
 
1 −1
C(Ũ ) = ρ(1) = cosh . (28)
2 2 ωr Re(ω(t))
This is the geodesic distance in the space of SL(2, R) unitaries with end points anchored at the
two points determined the boundary conditions (23).

4 Can Circuit Complexity Detect Chaos?

The goal of this paper is to explore whether we can implement the notion of quantum circuit com-
plexity as a diagnostic of a system’s chaotic behaviour. Classically, chaos is diagnosed by studying
trajectories in the phase space of some dynamical system., a notion that is not well definted in
quantum systems, essentially because of the uncertainty principle. It is important to keep in mind
then that when we speak of geodesics in the context of circuit complexity, we will mean trajectories
defined on the space of unitaries.

As a first step towards realising our goal, let us consider a quantum circuit responsible for
producing the state (4) from (2). We will then compute the complexity (geodesic distance) of the

5
 Figure 1: Complexity C(Ũ ) vs λ for t = 7

Figure 2: Complexity vs λ for different time. As t increases the other line is forming a cusp. It
seems the cusp is formed near t = 6.

target state (4) with respect to (2). It is evident from (5) and (28) that this complexity depends
both on time and coupling λ (defined in (1)). First we will study how this complexity changes with
coupling λ for a fixed time.

Interestingly, this displays a cusp-like singular behaviour. The cusp appears at the critical value
of λ = λc , where the frequency vanishes. Note that the system behaves like a regular oscillator for
λ < λc and an inverted oscillator for λ > λc . Since the inverted regime is unstable, it is natural
to expect that the complexity will increase with the increase of the coupling, as we would expect
the circuit complexity for an inverted oscillator will be large. This is precisely what we see in
Fig. (1), on top of the appearance of the cusp, the behaviour of complexity before and after λc
are completely different, as expected. After the critical value λc , when the system enters into the
unstable/chaotic regime, complexity grows rapidly with any further increase of the coupling λ.

To our surprise, complexity shows this cusp-like singular behaviour only after a certain time tc .
Within this time scale, i.e. for t < tc , the complexity behaves as a continuous function of coupling
λ with nothing special going on at λ = λc . But as we approach t = tc , the cusp starts to form.
To demonstrate this, in Fig. (2), we explore complexities for a set of gradually increasing values of

6
 Figure 3: Complexity vs time

time. For a particular choice of the parameter m, we can see that near t = 6 the cusp starts to
form. In other words, it takes a certain amount of time for the system to recognize the switch from
regular to inverted oscillator.

Before concluding this section, we explore the time evolution of the complexities for both regular
and inverted regimes. The results are displayed in Fig. (3). The complexity for the inverted
oscillator saturate very quickly. For the regular oscillator the complexity is oscillatory in nature
around some average value. Apart from the differences in their appearance, from this time evolution
we cannot make any precise statement for the chaotic nature of the system. In the following section
we will consider a different class of target state whose complexity will clearly capture the chaotic
behaviour.

5 Quantifying Chaos using Complexity

Now we propose a new diagnostic for chaotic behaviour based on circuit complexity. As discussed
above, the notion of “sensitive dependence on initial conditions” can be probed by evolving identical
states with slightly different Hamiltonians, Ĥ and Ĥ + δ Ĥ. Along these lines, consider a target
state |ψ2 i obtained by evolving a reference state |ψ0 i forward in time with Ĥ and then backward
in time with Ĥ + δ Ĥ,
|ψ2 i = ei(Ĥ+δĤ)t e−iĤt |ψ0 i . (29)
We would like to compute the complexity C( ˆ Ũ ) of this target state |ψ2 i with respect to the reference
state |ψ0 i [22]. For a chaotic quantum system, even if the two Hamiltonians Ĥ and Ĥ + δ Ĥ are
arbitrarily close, we will not be able to get arbitrarily close to the reference state |ψ0 i i.e |ψ2 i will
be quite different from |ψ0 i, even for δ Ĥ arbitrarily small.

Having done all the heavy lift already, from (28) we find [22]
 2 2 
ˆ Ũ ) = 1 cosh−1 ωr + |ω̂(t)|

C( , (30)
2 2 ωr Re(ω̂(t))
where now  
1 2
ψ2 (x, t) = N̂ (t) exp − ω̂(t)x , (31)
2

7
and
Ω02
ω̂(t) = i Ω0 cot(Ω0 t) + . (32)
sin2 (Ω0 t)(ω(t) + i Ω0 cot(Ω0 t))
√
In this last expression, Ω0 = m2 − λ0 is the frequency associated with the Hamiltonian H 0 = 21 p2 +
Ω02 2 0
2 x and λ = λ + δλ with δλ very small. The time dependence of this complexity demonstrates
that there is a clear qualitative difference between a regular oscillator and an inverted oscillator
as evident from Fig. (4) and Fig. (5). For the regular oscillator we get oscillatory behaviour
[22, 23, 24]; the complexity grows linearly for a very short period and reach to a saturation with
some fluctuations. However, C(ˆ Ũ ) for the inverted oscillator tells a completely different story.

ˆ Ũ ) vs time for Regular Oscillator (m = 1, λ = 1.2, δλ = 0.01)

Figure 4: C(

ˆ Ũ ) vs time for Inverted Oscillator (m = 1, λ = 15, δλ = 0.01)

Figure 5: C(

ˆ Ũ ) for the inverted oscillator appears to be some complicated mono-

The overall behaviour of C(
tonically growing function. However, a closer look at Fig. (5), reveals that it takes a small amount
of time for the complexity to pick up after which it displays a linear ramp with time. For a different
choice of coupling (λ > λc ) we get similar behaviour with different pick up time and slope (φ) for
the linear ramp. These features are displayed for different values of the coupling in Fig. (6).

As we increase λ (beyond the critical value), we are in effect making the model more unstable
and consequently from our very specific circuit model we expect a larger complexity and a smaller
pick up time. Therefore, the slope and pick up time scale are natural candidates for measuring the
unstable nature of the inverted oscillator. When we explore the slope φ of the linear region (as in

8
 ˆ Ũ ) vs time for different values of λ (with δλ = 0.01, m = 1)
Figure 6: C(

Fig. (6)) for different values of coupling λ we find the behaviour shown in Fig. (7). In the following
section, we will argue that this slope is similar to the Lyapunov exponent.

Figure 7: slope φ vs λ (δ = 0.01, m = 1)

Note that the linear growth kicks in near a certain time t = ts (as in the Fig. (5)) which depends
on the choice of the parameters, {m, λ}. We plot this pick up time as function of λ in Fig. (8). We
believe that this time scale is equivalent to the scrambling time which frequently appears in the
chaos literature. One way to confirm this is to compute the out-of-time-order four point correlator
(OTOC). The time when OTOC ∼ eΛ(t−t∗ ) becomes O(1), is called the scrambling time. For this
oscillator model (1), we can show it analytically following [42]. It is shown in the next section.

ˆ Ũ ) for the
We further check the sensitivity of our results to the magnitude of δλ. We plot the C(
inverted oscillator for a fixed value of λ but for different δλ. We find that while the slope of the
linear region remains same, the pick up time is sensitive to δλ as exhibited in Fig. (9).

9
 Figure 8: ts vs λ(δ = 0.01, m = 1)

ˆ Ũ ) on δλ (m = 1, λ = 10)
Figure 9: Dependence of C(

6 OTOC, Lyapunov Exponent and Scrambling Time

The exponential behaviour of the 4-point OTOC has recently emerged as a popular early-time
diagnostic for quantum chaos8 . In [42] the authors demonstrate explicit calculations of OTOCs for
harmonic oscillator. For our model, the OTOC for x and p operators (after reinstating the factor
of }) gives [42]
h[x(t), p]2 i = }2 cos2 Ω t, (33)
where Ω is defined (1). When Ω is imaginary, we can write the above expression as an exponential
function
h[x(t), p]2 i ≈ }2 e2 |Ω| t + · · · . (34)
Rewriting the above expression as e2λL (t−t∗ ) , with Lyapunov exponent λL allows us to immediately
read off that for our system, λL = |Ω| while the scrambling (or Ehrenfest) time is given by t∗ =
1 1 1
λL log } . The λ dependence of this time scale (in the units of log } ) is shown in Fig. 10.The nature
of the graph is in agreement with Fig. 8. In fact from Fig. 10 after doing a data-fitting we get for
An alternative to the OTOC, F (t) = hA† (t)B † (0)A(t)B(0)i, is the thermally averaged commutator-squared
8

C(t) = h[A(t), B(0)]2 i with the two being related through C(t) = 2 − 2Re(F (t)). Unless there is an explicit ambiguity,
we will refer to them both as the OTOC.

10
 Figure 10: t∗ vs λ (m=1).

the pick-up time,

4 log(2)
ts = . (35)
|Ω|
In the units of log }1 this is related to scrambling time t∗ as,

ts = 4 log(2)t∗ . (36)

Also, the λ dependence of the Lyapunov exponent is shown in Fig. 11. Again the nature of the
graph is in agreement with Fig. 7. After fitting the data we get for the slope φ of the linear region
of the graph in Fig. 7,
φ = 2|Ω| = 2 λL . (37)

Figure 11: λL vs λ (m=1).

We have also checked the m-dependence of the slope φ and the pick up time ts and they are in
agreement with the m-dependence of λL and t∗ respectively.

11
7 Towards a Field Theory Analysis
By using the single oscillator model we have illustrated how complexity can capture chaotic be-
haviour. In this section we will explore a possible field theory model in which the inverted oscillator
appears naturally. Consider two free scalar field theories ((1+1)-dimensional c = 1 conformal field
theories) deformed by a marginal coupling. The Hamiltonian is given by
Z Z
1 h
2 2 2 2 2 2 2
i
H = H0 + HI = dx Π1 + (∂x φ1 ) + Π2 + (∂x φ2 ) + m (φ1 + φ2 ) + λ dx(∂x φ1 )(∂x φ2 ).
2
(38)

We can discretize this theory by putting it on a lattice. Then using the following definitions
1 m
x(~n) = δφ(~n), p(~n) = Π(~n)/δ, ω = m, Ω = 2
, λ̂ = λ δ −4 and m̂ = , (39)
δ δ
we get
δ Xh 2 
H= p1,n + p22,n + Ω2 (x1,n+1 − x1,n )2 + Ω2 (x2,n+1 − x2,n )2 +
2 n
i (40)
2 2 2
m̂ (x1,n + x2,n ) + λ̂ (x1,n+1 − x1,n )(x2,n+1 − x2,n ) .

Next we perform a series of transformations,

N −1
1 X  2π i k 
x1,a = √ exp a x̃1,k ,
N k=0 N
N −1
1 X  2πik 
p1,a = √ exp − a p̃1,k ,
N k=0 N
N −1
1 X  2π i k 
x2,a = √ exp a x̃2,k ,
N k=0 N
N −1
1 X  2πik 
p2,a = √ exp − a p̃2,k ,
N k=0 N
ps,k + pa,k ps,k − pa,k
p̃1,k = √ , p̃2,k = √ ,
2 2
xs,k + xa,k xs,k − pa,k
x̃1,k = √ , x̃2,k = √ , (41)
2 2
that lead to the Hamiltonian
N −1
δ Xh 2 2 2 2 2 2
i
H= ps,k + Ω̄k xs,k + pa,k + Ωk xa,k , (42)
2
k=0

where   π k   
2 πk

Ω̄2k = 2 2
m̂ + 4 (Ω + λ̂) sin 2
, Ω2k 2 2
= m̂ + 4 (Ω − λ̂) sin . (43)
N N
It is immediately clear that by tuning the value of λ̂, the frequencies Ωk can be made arbitrarily
negative resulting in coupled inverted oscillators. Note that Ω̄k will be always positive. Therefore,

12
 Figure 12: C(Ũ ) vs λ for δ = 0.1, m = 1, t = 20, N = 1000.

one can view (42) as a sum of regular and inverted oscillator for each value of k. Now to study
the unstable behaviour, the regular oscillator part is not very interesting. Hence, we will simply
investigate the inverted oscillator part with the Hamiltonian
N −1     
δ X 2 2 2 2 πk
H̃(m, Ω, λ̂) = pk + m̂ + 4 (Ω − λ̂) sin x2k . (44)
2 N
k=0

Note that by tuning λ̂ for this Hamiltonian one can get both regular and inverted oscillators. Now
we want to time evolve with the above Hamiltonian. Specifically, at t = 0 we start with the ground
state of H̃(m, Ω, λ̂ = 0) and then time evolve it with H̃(m, Ω, λ̂ 6= 0). The complexity for this
time evolved target state with respect to the ground state of H̃(m, Ω, λ̂ = 0) is given by a suitable
generalization of (28)
v #!2
uN −1 "
2 + |ω (t)|2
1u X ωr,k k
C(Ũ ) = t cosh−1 , (45)
2 2 ωr,k Re(ωk (t))
k=0

where  
Ωk − i ωr,k cot(Ωk t)
ωk (t) = Ωk , (46)
ωr,k − i Ωk cot(Ωk t)
and π k
2
ωr,k = m̂2 + 4 Ω2 sin2 . (47)
N
We can also compute C( ˆ Ũ ) as before by considering two Hamiltonians H̃ and H̃ 0 with two slightly
different couplings, λ̂ and λ̂0 = λ̂ + δ λ̂, where δ λ̂ is small. In this case we get
v #!2
uN −1 "
2 + |ω̂ (t)|2
1 uX
−1
ωr,k k
ˆ Ũ ) =
C( t cosh , (48)
2 2 ωr,k Re(ω̂k (t))
k=0

where
Ω02 2 πk
 
ω̂k (t) = i Ω0k cot(Ω0k t) + k
, Ω02
k = m̂2
+ 4 (Ω2
− λ̂ − δ λ̂) sin
sin2 (Ω0k t) ωk (t) + i Ω0k cot(Ω0k t)


N
(49)

13
 ˆ Ũ ) vs time for the Inverted Oscillators (δ = 0.1, m = 1, N = 1000, λ̂δ 2 = 10, δλ = 0.01)
Figure 13: C(

and ωk (t)2 , ωr,k

2 are defined in (46).

Now we compute the complexity. Just as in the single oscillator case, we observe the formation
of a cusp as the system switches from regular to inverted oscillator. Fig. (12) shows the appearance
of cusp. Moreover, we confirmed that the cusp starts to appear at a certain time scale. Also, using
our testing method (outlined in section 5), once again for the inverted oscillator we can immediately
read off the scrambling time and Lyapunov exponent from the time evolution of C( ˆ Ũ ) as shown in
from Fig. (13).

8 Discussion
In this paper we used a harmonic oscillator model that converts to an inverted oscillator for large
coupling of the interaction Hamiltonian. The coupling behaves as a regulator and by tuning it we
can switch between regular and inverted regimes. Our motivation was to use this inverted oscillator
as a toy model to study quantum chaos. In this context, the regular oscillator serves as a reference
system. We developed a new diagnostic for quantum chaos by constructing a particular quantum
circuit and computing the corresponding complexity. Our diagnostic can extract equivalent infor-
mation as the out-of-time-order correlator with the additional feature that complexity can detect
when the system switches from regular to the chaotic regime.

By considering the target state as a time evolved state and investigating how complexity changes
with coupling, we find a new time scale, which we call the onset time. The onset time tells us how
long it takes for the system to recognize the transition from regular to unstable regime. Then we
considered a target state which is first forward evolved and then backward evolved with slightly
different Hamiltonians and found that the behaviour for the regular and inverted oscillator are
completely different in this case. For the regular oscillator we get some oscillatory behaviour as
in [21, 22, 24]. However, for the inverted oscillator we get an exponential type function with two
distinct features: for an initial period the complexity is nearly zero, after which it exhibits a steep
linear growth. By comparing with the operator product expansion, we discovered the small time
scale and slope of the linear portion to be equivalent to the scrambling time and the Lyapunov
exponent respectively.

14
 To give a proof-of-principle argument for complexity as a chaos diagnostic, we have used the
inverted oscillator as a toy model. This is, however, a rather special example and, by no means,
a realistic chaotic system. To put complexity on the same footing as, say the OTOC as a probe
of quantum chaos will take much more work, with more ‘realistic’ systems like the maximally-
chaotic SYK model9 and its many variants (see, for example, [44, 45, 46] and references therein)
in the (0+1)-dimensional quantum mechanical context, or the MSW class of (1+1)-dimensional
(non-maximally) chaotic conformal field theories [47].

As a final point of motivation, we note that by virtue of the recent ‘complexity=action’ [48]
and ‘complexity=volume’ [49] conjectures, the computational complexity of holographic quantum
system has a well-defined (if not entirely unambiguous) dual. This opens up tantalising new possi-
bilities in the study of quantum chaos in strongly coupled quantum systems. We leave these issues
for future work.

Acknowledgements
We would like to thank Dario Rosa and Jon Shock for useful comments. AB thanks Aninda Sinha,
Pratik Nandy, Jose Juan Fernandez-Melgarejo and Javier Molina Vilaplana for many discussions
and ongoing collaborations on complexity. AB also thanks Department of Physics, University of
Windsor, Perimeter Institute, Waterloo, FISPAC group, specially Jose Juan Fernandez-Melgarejo
and E. Torrente-Lujan, Department of Physics, University of Murcia for warm hospitality during
this work. AB is supported by JSPS Grant-in-Aid for JSPS fellows (17F17023). JM is supported
by the NRF of South Africa under grant CSUR 114599. NM is supported by the South African
Research Chairs Initiative of the Department of Science and Technology and the National Research
Foundation of South Africa. Any opinion, finding and conclusion or recommendation expressed
in this material is that of the authors and the NRF does not accept any liability in this regard.
Research at Perimeter Institute is supported by the Government of Canada through the Department
of Innovation, Science, and Economic Development, and by the Province of Ontario through the
Ministry of Research and Innovation.

References
[1] Interested readers are referred to the following thesis and the references therein.
N. Hunter-Jones, “ Chaos and Randomness in Strongly-Interacting Quantum Systems”, PhD the-
sis, California Institute of Technology, 2018, https://thesis.library.caltech.edu/11002/,

[2] F. Haake, “Quantum Signatures of Chaos,” Springer, 2010.

[3] A. Peres, Quantum theory: concepts and methods, Vol. 72 of Fundamental theories of physics
(Kluwer, Dordrecht, 1995).
9
It is worth pointing out that as this manuscript was being prepared, we were made aware of the recent article
[43] also advocing for the idea of complexity as a tool for the study of quantum chaotic systems.

15
[4] R. A. Jalabert and H. M. Pastawski, Environment-independent decoherence rate in classically
chaotic systems, Phys. Rev. Lett. 86, 2490 (2001).

[5] O. Bohigas, M. J. Giannoni, and C. Schmit, “Characterization of Chaotic Quantum Spectra

and Universality of Level Fluctuation Laws,” Phys. Rev. Lett. 52 (1984) 1.

[6] S. H. Shenker and D. Stanford, “Black holes and the butterfly effect,” JHEP 03 (2014) 067,
arXiv:1306.0622 [hep-th].

[7] S. H. Shenker and D. Stanford, “Stringy effects in scrambling,” JHEP 05 (2015) 132,
arXiv:1412.6087 [hep-th].

[8] A. Kitaev, “A simple model of quantum holography.” Talks given at the KITP, Apr. 7, 2015
and May 27, 2015.

[9] J. Maldacena, S. H. Shenker, and D. Stanford, “A bound on chaos,” JHEP 08 (2016) 106,
arXiv:1503.01409 [hep-th].

[10] A. I. Larkin and Y. N. Ovchinnikov, “Quasiclassical Method in the Theory of Superconduc-

tivity,” JETP 28 (1969) 1200.

[11] P. Hayden and J. Preskill, “Black holes as mirrors: Quantum information in random subsys-
tems,” JHEP 09 (2007) 120, arXiv:0708.4025 [hep-th].

[12] Y. Sekino and L. Susskind, “Fast Scramblers,” JHEP 10 (2008) 065, arXiv:0808.2096 [hep-th].

[13] N. Lashkari, D. Stanford, M. Hastings, T. Osborne, and P. Hayden, “Towards the Fast Scram-
bling Conjecture,” JHEP 04 (2013) 022, arXiv:1111.6580 [hep-th].

[14] T. Nosaka, D. Rosa and J. Yoon, JHEP 1809, 041 (2018) doi:10.1007/JHEP09(2018)041
[arXiv:1804.09934 [hep-th]].

[15] R. A. Jefferson and R. C. Myers, Circuit complexity in quantum field theory, JHEP 10 (2017)
107, [arXiv:1707.08570[hep-th]].

[16] M. Guo, J. Hernandez, R. C. Myers and S. M. Ruan, “Circuit Complexity for Coherent States,”
arXiv:1807.07677 [hep-th].

[17] L. Hackl and R. C. Myers, “Circuit complexity for free fermions,” arXiv:1803.10638 [hep-th]

[18] R. Khan, C. Krishnan, and S. Sharma, Circuit Complexity in Fermionic Field Theory,
[arXiv:1801.07620[hep-th]].

[19] S. Chapman, M. P. Heller, H. Marrochio and F. Pastawski, “Toward a Definition of Complexity

for Quantum Field Theory States,” Phys. Rev. Lett. 120 (2018) no.12, 121602 [arXiv:1707.08582
[hep-th]],

[20] H. A. Camargo, P. Caputa, D. Das, M. P. Heller and R. Jefferson, “Complexity as a novel

probe of quantum quenches: universal scalings and purifications,” arXiv:1807.07075 [hep-th],

16
[21] A. Bhattacharyya, A. Shekar and A. Sinha, “Circuit complexity in interacting QFTs and RG
flows,” JHEP 1810 (2018) 140 doi:10.1007/JHEP10(2018)140 [arXiv:1808.03105 [hep-th]].

[22] T. Ali, A. Bhattacharyya, S. Shajidul Haque, E. H. Kim and N. Moynihan, “Time Evolution of
Complexity: A Critique of Three Methods,” JHEP 1904 (2019) 087 [arXiv:1810.02734 [hep-th]].

[23] T. Ali, A. Bhattacharyya, S. Shajidul Haque, E. H. Kim and N. Moynihan, “Post-Quench

Evolution of Distance and Uncertainty in a Topological System: Complexity, Entanglement and
Revivals,” arXiv:1811.05985 [hep-th].

[24] S. Chapman, J. Eisert, L. Hackl, M. P. Heller, R. Jefferson, H. Marrochio and R. C. Myers,

“Complexity and entanglement for thermofield double states,” SciPost Phys. 6 (2019) no.3, 034
[arXiv:1810.05151 [hep-th]],

[25] K. Hashimoto, N. Iizuka, and S. Sugishita, Time Evolution of Complexity in Abelian

Gauge Theories - And Playing Quantum Othello Game, [arXiv:1707.03840[hep-th]], R.-Q.
Yang, A Complexity for Quantum Field Theory States and Application in Thermofield Double
States,[arXiv:1709.00921[hep-th]], R.-Q. Yang, C. Niu, C.-Y. Zhang, and K.-Y. Kim, Compari-
son of holographic and field theoretic complexities for time dependent thermofield double states,
JHEP 02 (2018) 082, [arXiv:1710.00600 [hep-th]], D. W. F. Alves and G. Camilo, “Evolution of
Complexity following a quantum quench in free field theory,” arXiv:1804.00107 [hep-th], J. Jiang,
J. Shan and J. Yang, ‘‘Circuit complexity for free Fermion with a mass quench,” arXiv:1810.00537
[hep-th], R. Q. Yang and K. Y. Kim, JHEP 1903 (2019) 010 [arXiv:1810.09405 [hep-th]], S. Liu,
“Complexity and scaling in quantum quench in 1 + 1 dimensional fermionic field theories,”
arXiv:1902.02945 [hep-th],

[26] P. Caputa, N. Kundu, M. Miyaji, T. Takayanagi and K. Watanabe, “Anti-de Sitter Space
from Optimization of Path Integrals in Conformal Field Theories,” Phys. Rev. Lett. 119 (2017)
no.7, 071602 [arXiv:1703.00456 [hep-th]], P. Caputa, N. Kundu, M. Miyaji, T. Takayanagi and
K. Watanabe, “Liouville Action as Path-Integral Complexity: From Continuous Tensor Networks
to AdS/CFT,” JHEP 1711 (2017) 097 [arXiv:1706.07056 [hep-th]], B. Czech, “Einstein Equations
from Varying Complexity,” Phys. Rev. Lett. 120 (2018) no.3, 031601 [arXiv:1706.00965 [hep-th]],
J. Molina-Vilaplana and A. Del Campo, “Complexity Functionals and Complexity Growth Lim-
its in Continuous MERA Circuits,” arXiv:1803.02356 [hep-th], A. Bhattacharyya, P. Caputa,
S. R. Das, N. Kundu, M. Miyaji and T. Takayanagi, “Path-Integral Complexity for Perturbed
CFTs,” JHEP 1807 (2018) 086 [arXiv:1804.01999 [hep-th]], T. Takayanagi, “Holographic Space-
times as Quantum Circuits of Path-Integrations,” arXiv:1808.09072 [hep-th], H. A. Camargo,
M. P. Heller, R. Jefferson and J. Knaute, “Path integral optimization as circuit complexity,”
arXiv:1904.02713 [hep-th].

[27] R. Q. Yang, Y. S. An, C. Niu, C. Y. Zhang and K. Y. Kim, “Axiomatic complexity in quan-
tum field theory and its applications,” arXiv:1803.01797 [hep-th], P. Caputa and J. M. Ma-
gan, “Quantum Computation as Gravity,” arXiv:1807.04422 [hep-th], R. Q. Yang, Y. S. An,
C. Niu, C. Y. Zhang and K. Y. Kim, “More on complexity of operators in quantum field theory,”
arXiv:1809.06678 [hep-th], A. Bernamonti, F. Galli, J. Hernandez, R. C. Myers, S. M. Ruan
and J. Simn, “The First Law of Complexity,” arXiv:1903.04511 [hep-th], I. Akal, “Weighting

17
 gates in circuit complexity and holography,” arXiv:1903.06156 [hep-th] G. Camilo, D. Melnikov,
F. Novaes and A. Prudenziati, “Circuit Complexity of Knot States in Chern-Simons theory,”
arXiv:1903.10609 [hep-th] A. R. Brown and L. Susskind, “The Complexity Geometry of a Single
Qubit,” arXiv:1903.12621 [hep-th], D. Ge and G. Policastro, “Circuit Complexity and 2D Boson-
isation,” arXiv:1904.03003 [hep-th] W. H. Huang, “Operator Approach to Complexity : Excited
State,” arXiv:1905.02041 [hep-th], V. Balasubramanian, M. DeCross, A. Kar and O. Parrikar,
“Binding Complexity and Multiparty Entanglement,” JHEP 1902 (2019) 069 [arXiv:1811.04085
[hep-th]].

[28] M. A. Nielsen, A geometric approach to quantum circuit lower bounds, [arXiv:quant-

ph/0502070].

[29] M. A. Nielsen, M. R. Dowling, M. Gu, and A. M. Doherty, Quantum Computation as Geometry,

Science 311 (2006) 1133–1135, [arXiv:quant-ph/0603161].

[30] M. A. Nielsen and M. R. Dowling, The geometry of quantum computation, [arXiv:quant-

ph/0701004].

[31] M. Miyaji, “Butterflies from Information Metric,” JHEP 1609 (2016) 002
doi:10.1007/JHEP09(2016)002 [arXiv:1607.01467 [hep-th]].

[32] P. Hosur, X.-L. Qi, D. A. Roberts, and B. Yoshida, “Chaos in quantum channels,” JHEP 02
(2016) 004, arXiv:1511.04021 [hep-th].

[33] D. A. Roberts and B. Yoshida, “Chaos and complexity by design,” JHEP 04 (2017) 121,
arXiv:1610.04903 [quant-ph].

[34] J. Cotler, N. Hunter-Jones, J. Liu, and B. Yoshida, “Chaos, Complexity, and Random Matri-
ces,” JHEP 11 (2017) 048, arXiv:1706.05400 [hep-th].

[35] J. M. Magan, “Black holes, complexity and quantum chaos,” arXiv:1805.05839 [hep-th].

[36] V. Balasubramanian, M. DeCross, A. Kar and O. Parrikar, “Quantum Complexity of Time

Evolution with Chaotic Hamiltonians,” arXiv:1905.05765 [hep-th].

[37] R. Blume-Kohout and W. Zurek, Decoherence from a chaotic environment: an upside-down

“oscillator” as a model, Phys. Rev. A 68, 032104 (2003).

[38] G. Barton, Quantum mechanics of the inverted oscillator, Ann. Phys. 166, 322 (1986).

[39] S. Gentilini, M. C. Braidotti, G. Marcucci, E. DelRe, and C. Conti, Physical realization of the
Glauber quantum oscillator, Sci. Rep. 5, 15816 (2015).

[40] M. V. Berry and J. P. Keating, The Riemann zeros and eigenvalue asymptotics, SIAM Rev.
41, 236 (1999).

[41] See, e.g. R. Shankar, Principles of Quantum Mechanics (Plenum, New York, 1994).

[42] K. Hashimoto, K. Murata, and R. Yoshii: Out-of-time-order correlators in quantum mechanics,

JHEP 10 138 (2017).

18
[43] V. Balasubramanian, M. DeCross, A. Kar and O. Parrikar, arXiv:1905.05765 [hep-th].

[44] J. Maldacena and D. Stanford, Phys. Rev. D 94, no. 10, 106002 (2016)
doi:10.1103/PhysRevD.94.106002 [arXiv:1604.07818 [hep-th]].

[45] W. Fu, D. Gaiotto, J. Maldacena and S. Sachdev, Phys. Rev. D 95, no. 2, 026009 (2017) Adden-
dum: [Phys. Rev. D 95, no. 6, 069904 (2017)] doi:10.1103/PhysRevD.95.069904, 10.1103/Phys-
RevD.95.026009 [arXiv:1610.08917 [hep-th]].

[46] A. Kitaev and S. J. Suh, JHEP 1805, 183 (2018) doi:10.1007/JHEP05(2018)183

[arXiv:1711.08467 [hep-th]].

[47] J. Murugan, D. Stanford and E. Witten, JHEP 1708, 146 (2017)

doi:10.1007/JHEP08(2017)146 [arXiv:1706.05362 [hep-th]].

[48] A. R. Brown, D. A. Roberts, L. Susskind, B. Swingle and Y. Zhao, Phys. Rev. Lett. 116, no.
19, 191301 (2016) doi:10.1103/PhysRevLett.116.191301 [arXiv:1509.07876 [hep-th]].

[49] D. A. Roberts, D. Stanford and L. Susskind, JHEP 1503, 051 (2015)

doi:10.1007/JHEP03(2015)051 [arXiv:1409.8180 [hep-th]].

19