Concepts of Quantum

Computation
G.P. Biswas
Prof./CSE, IIT(ISM), Dhanbad
Introduction
• It is said that
• The nineteenth century was known as the Machine Age, the twentieth century
has gone down in history as the Information Age. It is believed that the twenty-
first century will be the Quantum Age.

—Paul Davies, Professor Natural Philosophy – Australian Centre for

Astrobiology

• Quantum mechanics, that mysterious, confusing discipline, which

none of us really understands, but which we know how to use.
—Murray Gell-Mann
Contd.
• Quantum computing is a beautiful combination of quantum physics, computer
science, and information theory.
• Information theory (ICT), which includes the foundations of both computer
science and communications, abstracted away the physical world so effectively
that it became possible to talk about the major issues within computer science
and communications, such as the efficiency of an algorithm or the robustness of a
communication protocol, without understanding details of the physical devices
used for the computation or the communication.
• Quantum mechanics (relates to Quantum physics and Quantum Computing) has
played an ever-increasing role in the development of new and more efficient
computing devices. Quantum mechanics underlies the working of traditional,
classical computers and communication devices, from the transistor through the
laser to the latest hardware advances that increase the speed and power and
decrease the size of computer and communications components. Until recently,
the influence of quantum mechanics remained confined to the low level
implementation realm; it had no effect on how computation or communication
was thought of or studied.
Contd.
• Quantum information processing, a field that includes quantum
computing, quantum cryptography, quantum communications, and
quantum games, explores the implications of using quantum mechanics
instead of classical mechanics to model information and its processing.
• Quantum computing is not about changing the physical substrate on which
computation is done from classical to quantum, but rather changing the
notion of computation itself.
• The change starts at the most basic level:
• Fundamental unit of computation is no longer the bit, but rather the quantum bit
or qubit.
• Placing computation on a quantum mechanical foundation led to the
discovery of faster algorithms, novel cryptographic mechanisms, and
improved communication protocols.
Contd.
• The phrase quantum computing does not parallel the phrases DNA
computing or optical computing: these describe the substrate on
which computation is done without changing the notion of
computation.
• Classical computers, the ones we all have on our desks, make use of
quantum mechanics, but they compute using bits, not qubits. For this
reason, they are not considered quantum computers.
• A quantum or classical computer may or may not be an optical
computer, depending on whether optical devices are used to carry
out the computation. Whether the computer is quantum or classical
depends on whether the information is represented and manipulated
in a quantum or classical way.
Contd.
• The quantum computing is closer in character to analog computing
because the computational model for analog computing differs from that
of standard computing: a continuum of values, rather than only a discrete
set, is allowed.
• While the phrases are parallel, the two models differ greatly in that analog
computation does not support entanglement, a key resource for quantum
computation, and measurements of a quantum computer’s registers can
yield only a small, discrete set of values.
• Furthermore, while a qubit can take on a continuum of values, in many
ways a qubit resembles a bit, with its two discrete values, more than it
does analog computation (for example, only one bit’s worth of information
can be extracted from a qubit by measurement.
What is a quantum computer?

• A quantum computer is a machine that performs calculations based on

the laws of quantum mechanics, which is the behavior of particles at
the sub-atomic level:

• Quantum mechanics is a more fundamental theory than Newtonian

mechanics and classical electromagnetism
• It provides accurate and precise descriptions for many phenomena that these
"classical" theories simply cannot explain on the atomic and subatomic level
An Illustration of Quantum Computation with a simple experiment (in optics)

Consider a setup involving a photon source, a half-silvered

mirror (beam splitter), and a pair of photon detectors.

detectors

photon
beamsplitter
source
Now consider what happens when a single photon is
fired?

50%

50%

Simplest explanation: Beam-splitter acts as a

classical coin-flip, randomly sending each
photon one way or the other.
 Classical probabilities

Consider a computation tree for a simple two-step (classical) probabilistic algorithm,

which makes a coin-flip at each step, and whose output is 0 or 1:
The probability of the computation following a
1
2
0 given path is obtained by multiplying the
probabilities along all branches of that path… in
1
2 the example the probability the computation
1
1 follows the red path is
2
1 1 1
1
2
0  
2 2 4
The probability of the computation giving the
1 answer 0 is obtained by adding the probabilities
2
1
of all paths resulting in 0:
1 1 1
 
4 4 2
 The “weirdness” of quantum mechanics

Consider a modification of the experiment-

The simplest explanation for the 100%
modified setup would still
predict a 50-50 distribution

full mirror

The simplest explanation is wrong!

 vs quantum probabilities

In quantum physics, we have probability amplitudes, which can

have complex phase factors associated with them.
1
|0 The probability amplitude associated with a path in the
2 computation tree is obtained by multiplying the
1 probability amplitudes on that path. In the example,
2 1 the red path has amplitude 1/2, and the green path
2
|1 has amplitude –1/2.
1 |0
 The probability amplitude for getting the answer |0 is
1 2
obtained by adding the probability amplitudes… notice
2
1
that the phase factors can lead to cancellations! The
2 |1 probability of obtaining |0 is obtained by squaring the
total probability amplitude. In the example the
probability of getting |0 is
2
1 1
   0
2 2
 Explanation of experiment

Consider a modification of the experiment…

The simplest explanation for the 100%
modified setup would still 1 1
0  0 1 0
predict a 50-50 distribution… 2 2

1 1 1
1 1  1  01
2 2 2
1 full mirror
0 0
2
Realization of Single (Quantum) Qubit System
• Quantum bits (qubit) are the fundamental units of information in
quantum information processing in much the same way that bits are
the fundamental units of information for classical processing.
• Just as there are many ways to realize classical bits physically (two
voltage levels, lights on or off , positions of toggle switches), there are
many ways to realize quantum bits physically.
• As is done in classical computer science, we will concern ourselves
only rarely with how the quantum bits are realized. For the concretely
illustrating quantum bits and their properties, the behavior of
polarized photons, one of many possible realizations
of quantum bits, may be looked at.
Two Slit Experiment
• Richard Feynman said that the two-slit experiment contains “all of the
mystery” of quantum mechanics.
• It is shown that the two-slit experiment does contain the key mystery
of quantum mechanics, but it’s not exactly what Feynman described
in 1964. It is connected to the new mystery of “entanglement.”
• Feynman’s mystery was simply how a particle can interfere with itself
if it goes through only one slit. Our view is that the particle goes
through one slit. We show that it is the probability amplitude of the
wave function that is interfering with itself.
• you look at it the more mysterious it seems.” “How can such an
interference come about?”, he asks. “Perhaps...it is not true that the
lumps go either through hole 1 or hole 2.” He says,
• We conclude the following: The electrons arrive in lumps, like particles, and
the probability of arrival of these lumps is distributed like the distribution of
intensity of a wave. It is in this sense that an electron behaves “sometimes
like a particle and sometimes like a wave”... The only answer that can be given
is that we have found from experiment that there is a certain special way that
we have to think in order that we do not get into inconsistencies. What we
must say (to avoid making wrong predictions) is the following. If one looks at
the holes or, more accurately, if one has a piece of apparatus which is capable
of determining whether the electrons go through hole 1 or hole 2, then one
can say that it goes either through hole 1 or hole 2. But, when one does not
try to tell which way the electron goes, when there is nothing in the
experiment to disturb the electrons, then one may not say that an electron
goes either through hole 1 or hole 2. If one does say that, and starts to make
any deductions from the statement, he will make errors in the analysis. This is
the logical tightrope on which we must walk if we wish to describe nature
successfully.
Quantum Mechanics of Photon Polarization
• It’s simple experiment that illustrates some of the nonintuitive
behavior of quantum systems, behavior exploited to good effect in
quantum algorithms and protocols. This experiment can be
performed by using only minimal equipment:
• a laser pointer and three polaroids (polarization filters), which are readily
available (from any camera supply store).
• The formalisms of quantum mechanics that describe this simple
experiment lead directly to a description of the quantum bit, the
fundamental unit of quantum information on which quantum
information processing is done.
• The experiment not only gives a concrete realization of a quantum
bit, but it also illustrates key properties of quantum measurement.
The Experiment
• Shine a beam of light on a projection screen as shown in the figure. When polaroid A is placed
between the light source and the screen, the intensity of the light reaching the screen is reduced.
Let us suppose that the polarization of polaroid A is horizontal.
• Next, place polaroid C between polaroid A and the projection screen. If polaroid C is rotated so
that its polarization is orthogonal (vertical) to the polarization of A, no light reaches the screen.
Contd.
• Finally, place polaroid B between polaroids A and C. One might expect that adding another
polaroid will not make any difference; if no light got through two polaroids, then surely no light
will pass through three! Surprisingly, at most polarization angles of B, light shines on the screen.
• The intensity of this light will be maximal if the polarization of B is at 45 degrees to both A and C.
Clearly the polaroids cannot be acting as simple sieves; otherwise, inserting polaroid B could not
increase the number of photons reaching the screen.
Quantum Explanation of the Experiment
• For a bright beam of light, there is a classical explanation of the experiment
in terms of waves. Versions of the experiment described here, using light so
dim that only one photon at a time interacts with the polaroid, have been
done with more sophisticated equipment.
• The results of these single photon experiments can be explained only using
quantum mechanics (the classical wave explanation no longer works).
• Furthermore, it is not just light that behaves in this peculiar way. The
quantum mechanical explanation of the experiment consists of two parts:
• a model of a photon’s polarization state and a model of the interaction between a
polaroid and a photon.
• The description of this experiment, and the definition of a qubit, use basic
notions of linear algebra such as vector, basis, orthonormal, and linear
combination.
Contd.
• Quantum mechanics models a photon’s polarization state by a unit vector, a vector of length 1,
pointing in the appropriate direction.
• We write |↑> and |→> for the unit vectors that represent vertical and horizontal polarization
respectively. Think of |v> as a vector with some arbitrary label v.
• In quantum mechanics, the standard notation for a vector representing a quantum state is |v>
(just as v or v are notations used for vectors in other settings). This notation is part of a more
general notation, Dirac’s notation, An arbitrary polarization can be expressed as a linear
combination |v >= a|↑> + b|→> of the two basis vectors |↑> and |→>.
Contd.
Contd.
• Quantum mechanics models the interaction between a photon and a polaroid as follows.
• The polaroid has a preferred axis, its polarization. When a photon with polarization |v>=
a|↑> + b|→> meets a polaroid with preferred axis |↑>, the photon will get through with
probability |a|^2 and will be absorbed with probability |b|^2; the probability that a photon
passes through the polaroid is the square of the magnitude of the amplitude of its
polarization in the direction of the polaroid’s preferred axis.
• The probability that the photon is absorbed by the polaroid is the square of the magnitude of
the amplitude in the direction perpendicular to the polaroid’s preferred axis.
• Furthermore, any photon that passes through the polaroid will now be polarized in the direction
of the polaroid’s preferred axis. The probabilistic nature of the interaction and the resulting
change of state are features of all interactions between qubits and measuring devices, no matter
what their physical realization.
• In the experiment, any photons that pass through polaroid A will leave polarized in the direction
of polaroid A’s preferred axis, in this case horizontal, |→>. A horizontally polarized photon has
no amplitude in the vertical direction, so it has no chance of passing through polaroid C, which
was given a vertical orientation. For this reason, no light reaches the screen. Had polaroid C
been in any other orientation, a horizontally polarized photon would have some amplitude in the
direction of polaroid C’s preferred axis, and some photons would reach the screen.
Contd.
Single Quantum Bit
• The space of possible polarization states of a photon is an example of a quantum bit, or qubit. A
qubit has a continuum of possible values:
• Any state represented by a unit vector a|↑> + b|→> is a legitimate qubit value. The amplitudes a and b can
be complex numbers, even though complex amplitudes were not needed for the explanation of the
experiment. (In the photon polarization case, the imaginary coefficients correspond to circular polarization.)
• In general, the set of all possible states of a physical system is called the state space of the system.
Any quantum mechanical system that can be modeled by a two-dimensional complex vector
space can be viewed as a qubit.
• Such systems, called two state quantum systems, include photon polarization, electron spin, and
the ground state together with an excited state of an atom. The two-state label for these systems
does not mean that the state space has only two states—it has infinitely many—but rather that
all possible states can be represented as a linear combination, or superposition, of just two states.
• For a two-dimensional complex vector space to be viewed as a qubit, two linearly independent
states, labeled |0> and |1>, must be distinguished. For the theory of quantum information
processing, all two-state systems, whether they be electron spin or energy levels of an atom, are
equally good.
• From a practical point of view, it is as yet unclear which two-state systems will be most suitable
for physical realizations of quantum information processing devices such as quantum computers;
it is likely that a variety of physical representation of qubits will be used.
Contd.
• Dirac’s bra / ket notation is used (throughout quantum physics) to represent quantum states and
their transformations.

set of vectors S, the subspace of all linear combinations of vectors in S is called the span of S and
is denoted span(S). A set of vectors B for which every element of V can be written uniquely as a
linear combination of vectors in B is called a basis for V . In a two-dimensional vector space, any
two vectors that are not multiples of each other form a basis. In quantum mechanics, bases are
usually require to be orthonormal. The two distinguished states, |0 >and |1>, are also required to
be orthonormal.
Contd.
Contd.
Contd.
A Quantum Key Distribution Protocol
• Keys—binary strings or numbers chosen randomly from a sufficiently large
set—provide the security for most cryptographic protocols, from
encryption to authentication to secret sharing. For this reason, the
establishment of keys between the parties who wish to communicate is of
fundamental importance in cryptography.
• Two general classes of keys exist:
• symmetric keys and public-private key pairs. Both types are used widely, often in
conjunction, in a wide variety of practical settings, from secure e-commerce
transactions to private communication over public networks.
• Public-private key pairs consist of a public key, knowable by all, and a corresponding
private key whose secrecy must be carefully guarded by the owner. Symmetric keys
consist of a single key (or a pair of keys easily computable from one another) that are
known to all of the legitimate parties and no one else. In the symmetric key case,
multiple parties are responsible for guarding the security of the key
Contd.
• Quantum key distribution protocols establish a symmetric key
between two parties, who are generally known in the cryptographic
community as Alice and Bob.
• Quantum key distribution protocols can be used securely anywhere
classical key agreement protocols such as Diffie-Hellman can be used.
They perform the same task; however, the security of quantum key
distribution rests on fundamental properties of quantum mechanics,
whereas classical key agreement protocols rely on the computational
intractability of a certain problem.
• For example, while Diffie-Hellman remains secure against all known classical
attacks, the problem on which it is based, the discrete logarithm problem
(DLP), is tractable on a quantum computer.
BB84 Protocol
• The earliest quantum key distribution protocol is known as BB84 after
its inventors, Charles Bennett and Gilles Brassard, and the year of the
invention.
• The aim of the BB84 protocol is to establish a secret key, a random
sequence of bit values 0 and 1, known only to the two parties, Alice
and Bob, who may use this key to support a cryptographic task such
as exchanging secret messages or detecting tampering.
• The BB84 protocol enables Alice and Bob to be sure that if they
detect no problems while attempting to establish a key, then with
high probability it is secret. The protocol does not guarantee,
however, that they will succeed in establishing a private key
Contd.
• Suppose Alice and Bob are connected by two public channels:
• an ordinary bidirectional classical channel and a unidirectional quantum channel. The
quantum channel allows Alice to send a sequence of single qubits to Bob (suppose the qubits
are encoded in the polarization states of individual photons).
• Both channels can be observed by an eavesdropper Eve as shown in figure.
Contd.
• To begin the process of establishing a private key, Alice uses quantum
or classical means to generate a random sequence of classical bit values and a random
subset of this sequence will be the final private key.
• Alice then randomly encodes each bit of this sequence in the polarization state of a
photon by randomly choosing for each bit one of the following two agreed-upon bases in
which to encode it: the standard basis are

• She sends this sequence of photons to Bob through the quantum channel
Contd.
• Bob measures the state of each photon he receives by randomly picking
either basis.
• Over the classical channel, Alice and Bob check that Bob has received a
photon for every one Alice has sent, and only then do Alice and Bob tell
each other the bases they used for encoding and decoding (& measuring)
each bit.
• When the choice of bases agree, Bob’s measured bit value agrees with the
bit value that Alice sent. When they chose different bases, the chance that
Bob’s bit matches Alice’s is only 50 percent. Without revealing the bit
values themselves, which would also reveal the values to Eve, there is no
way for Alice and Bob to figure out which of these bit values agree and
which do not. So they simply discard all the bits on which their choice of
bases differed. An average of 50 percent of all bits transmitted remain.
Then, depending on the level of assurance they require, Alice and Bob
compare a certain number of bit values to check that no eavesdropping has
occurred. These bits will also be discarded, and only the remaining bits will
be used as their private key.
Analysis
• Now, we describe one sort of attack that Eve can make and how quantum
aspects of this protocol guard against it.
• On the classical channel, Alice and Bob discuss only the choice of bases and
not the bit values themselves, so Eve cannot gain any information about
the key from listening to the classical channel alone.
• To gain information, Eve must intercept the photons transmitted by Alice
through the quantum channel. Eve must send photons to Bob before
knowing the choice of bases made by Alice and Bob, because they compare
bases only after Bob has confirmed receipt of the photons.
• If she sends different photons to Bob, Alice and Bob will detect that
something is wrong when they compare bit values, but if she sends the
original photons to Bob without doing anything, she gains no information.
Contd.
• Also note that since Alice has not yet told Bob her sequence of bases, Eve does
not know in which basis to measure each bit. If she randomly measures the bits,
she will measure using the wrong basis approximately half of the time. When she
uses the wrong basis to measure, the measurement changes the polarization of
the photon before it is resent to Bob. This change in the polarization means that,
even if Bob measures the photon in the same basis as Alice used to encode the
bit, he will get the correct bit value only half the time.
• Overall, for each of the qubits Alice and Bob retain, if the qubit was measured by
Eve before she sent it to Bob, there will be a 25 percent chance that Bob
measures a different bit value than the one Alice sent. Thus, this attack on the
quantum channel is bound to introduce a high error rate that Alice and Bob
detect by comparing a sufficient number of bits over the classical channel.
• If these bits agree, they can confidently use the remaining bits as their private
key. So, not only is it likely that 25 percent of Eve’s version of the key is incorrect,
but the fact that someone is eavesdropping can be detected by Alice and Bob.
Thus Alice and Bob run little risk of establishing a compromised key; either they
succeed in creating a private key or they detect that eavesdropping has taken
place.
Contd.
• Eve does not know in which basis to measure the qubits, a property crucial to the
security of this protocol, because Alice and Bob share information about which
bases they used only after Bob has received the photons.
• If Eve knew in which basis to measure the photons, her measurements would not
change the state, and she could obtain the bit values without Bob and Alice
noticing anything suspicious. A seemingly easy way for Eve to overcome this
obstacle is for her to copy the qubit, keeping a copy for herself while sending the
original on to Bob. Then she can measure her copy later after learning the correct
basis from listening in on the classical channel.
• Such a protocol is defeated by an important property of quantum information. As
we know that the no-cloning principle of quantum mechanics means that it is
impossible to reliably copy quantum information unless a basis in which it is
encoded is known; all quantum copying machines are basis dependent. Copying
with the wrong machine not only does not produce an accurate copy, but it also
changes the original in much the same way measuring in the wrong basis does. So
Bob and Alice would detect attempts to copy with high probability.
Multiple-Qubit Systems
• Unlike classical systems, the state space of a quantum system grows
exponentially with the number of particles. Thus, when we encode
computational information in quantum states of a system of n particles,
there are vastly more possible computation states available than when
classical states are used to encode the information.
• The extent to which these large state spaces corresponding to small
amounts of physical space can be used to speed up computation.
• The enormous difference in dimension between classical and quantum
state spaces is due to a difference in the way the spaces combine (Imagine
a macroscopic physical system consisting of several components. The state
of this classical system can be completely characterized by describing the
state of each of its component pieces separately).
• A surprising and unintuitive aspect of quantum systems is that often the
state of a system cannot be described in terms of the states of its
component pieces. States that cannot be so described are called entangled
states. Entangled states are a critical ingredient of quantum computation.
Contd.
• Entangled states are a uniquely quantum phenomenon; they have no
classical counterpart. Most states in a multiple-qubit system are entangled
states; they are what fills the vast quantum state spaces.
• The impossibility of efficiently simulating the behavior of entangled states
on classical computers suggested to Feynman, Manin, and others that it
might be possible to use these quantum behaviors to compute more
efficiently, leading to the development of the field of quantum
computation.
• In classical physics, the possible states of a system of n objects, whose
individual states can be described by a vector in a two-dimensional vector
space, can be described by vectors in a vector space of 2n dimensions.
Classical state spaces combine through the direct sum.
• However, the combined state space of n quantum systems, each with
states modeled by two-dimensional vectors, is much larger. The vector
spaces associated with the quantum systems combine through the tensor
product, resulting in a vector space of 2^n dimensions.
Tensor Products of Vector Spaces
The State Space of an n-Qubit System
Basic Concepts of Quantum Computation (Qubits, Registers,Gates and Networks)
• Consider two binary strings: 011 and 111. The first one can represent, for example, the number 3
(in binary) and the second one the number 7. In general three physical bits can be prepared in
2^3 = 8 different configurations that can represent, for example, the integers from 0 to 7.
However, a register composed of three classical bits can store only one number at a time.
• Consider qubits and quantum registers:
• A qubit is a quantum system in which the Boolean states 0 and 1 are represented by a prescribed pair of
normalized and mutually orthogonal quantum states labeled as {|0>, |1>}. The two states form a
‘computational basis’ and any other (pure) state of the qubit can be written as a superposition α|0>+β|1> for
some α and β such that |α|^2 + |β|^2 = 1. A qubit is typically a microscopic system, such as an atom, a
nuclear spin, or a polarized photon. A collection of n qubits is called a quantum register of size n.
• In quantum registrar, a number 6 (say) is represented by |1 >⊗ |1 >⊗ |0 >. In more compact
notation: |a> stands for the tensor product |𝑎𝑛−1 >⊗ |𝑎𝑛−2 > ⋯ |𝑎1 >⊗ |𝑎0 0>, where
∈ {0, 1} and it represent a quantum registrar prepared with the value 𝑎 = 2 𝑎0 + 21 𝑎1 + ⋯ +
𝑎1𝑛−1
2 𝑎𝑛−1 .
• There are 2^n states of this kind, representing all binary strings of length n or numbers from 0 to
2^n - 1, and they form a convenient computational basis. In the following a ∈ {0, 1}^n
(a is a binary string of length n) implies that |a> belongs to the computational
basis.
• But, it can also store the two of them simultaneously: Take first qubit and represent as a
superposition
Unitary Operators
• These preparations, and any other manipulations on qubits, have to be performed by unitary
operations. A quantum logic gate is a device which performs a fixed unitary operation on selected
qubits in a fixed period of time and a quantum network is a device consisting of quantum logic
gates whose computational steps are synchronized in time. The outputs of some of the gates are
connected by wires to the inputs of others. The size of the network is the number of gates it
contains.
Postulates of Quantum Mechanics
• An important distinction needs to be made between quantum
mechanics, quantum physics and quantum computing.
• Quantum mechanics is a mathematical language, much like calculus, just
as classical physics uses calculus to explain nature.
• Quantum physics uses quantum mechanics to explain nature. Just as classical
computers can be thought of in Boolean algebra terms,
• Quantum computers are reasoned about with quantum mechanics.
• There are four postulates to quantum mechanics, which will form the
basis of quantum computers:
• Postulate 1: Definition of a quantum bit, or qubit.
• Postulate 2: How qubit(s) transform (evolve).
• Postulate 3: The effect of measurement.
• Postulate 4: How qubits combine together into systems of qubits
A Quantum bit
• Postulate 1 (Nielsen and Chuang, page 80):

“Associated to any isolated physical system is a complex vector space with inner product
(i.e. a Hilbert space) known as the state space of the system. The system is
completely described by its state vector, which is a unit vector in the system’s state
space.”
• Consider a single qubit - a two-dimensional state space. Let |𝜙0 > and |𝜙1 > will be
orthonormal basis for the space. Then a qubit |𝜓 > = 𝑎|𝜙0 >+b |𝜙1 >.
• In quantum computing, we usually label the basis with some boolean name but note
carefully that this is only a name. For example, |𝜙0 >= |0> [(1 0)^t and |𝜙0 >=|1> (0
1)^t]. Making this more concrete one might imagine that |0> is being represented by an
up-spin while |1> by a down-spin. The key is there is an abstraction between the
technology (spin state or other quantum phenomena) and the logical meaning. This
same detachment occurs classically where we traditionally call a high positive voltage “1”
and a low ground potential “0”.
• Note that
2
|𝜓 2> = 𝑎|𝜙0 >+b |𝜙1 > must be a unit vector. In other words, < 𝜓 |𝜓 > = 1,
or |𝑎| + |𝑏| = 1, for quantum computing {𝑎, 𝑏} ∈ 𝐶 (C stands for the set of complex
numbers).
Contd.
• This formalism for a quantum bit is a direct extension of one way to
describe a classical computer. That is, the way we may write that a
classical bit 𝜔 > = 𝑥 0 > +𝑦|1 >. The only difference is x and y are
defined not over the complex numbers but rather from the set {0, 1}.
The same normalization condition applies |𝑥|2 + |𝑦|2 = 1, however,
this normalization condition is not a property of quantum mechanics
but rather of probability theory.
Ket Notation (|>)
• We introduce a bit of unusual notation that will play an essential role
in quantum computation. This is the “ket” notation invented by Paul
Dirac, and thus, known as Dirac notation (|>).
• Actually, we will enclose bits and bit-strings in asymmetrical brackets
called kets, writing|0〉and|1〉instead of 0 and 1 (in classical). We
will also usually eliminate internal brackets when writing strings; e.g.,
writing|011〉instead of |0〉|1> |1>.
Evolution of quantum systems (Postulate 2)
“The evolution of a closed quantum system is described by a unitary transformation. That is, the
state |𝜓 > of the system at time 𝑡1 is related to the state of |𝜓′ > of the system at time 𝑡2 by a
unitary operator U which depends only on times 𝑡1 and 𝑡2 ”, i.e. |𝜓′ >=U |𝜓 >.
• The fact that U cannot depend on |𝜓 > and only on 𝑡1 and 𝑡2 (is a subtle and disappointing fact)
(if U could depend on |𝜓 >, then quantum computers could easily solve NP complete problems!
• Conceptually think of U as something you can apply to a quantum bit but you cannot
conditionally apply it. The transform occurs without any regard to the current state of |𝜓 >.
0 1
• For example, consider |𝜓 > = a|0> + b|1> and 𝑈 = , then
0 1 𝑎 𝑏 1 0
′
|𝜓 > = 𝑈 |𝜓 > = = = b|0> + a|1>
1 0 𝑏 𝑎
Contd.
Reversible Computation
• In actual physical reality, a theoretical bit (|0〉or|1〉) is implemented by a particle or bunch of
particles (e.g., high or low voltage on a physical wire).
• Similarly, a gate is implemented by a physical object (a “switch” or some other gadget) that
manipulates the bit-representations.
• We then would ideally like to think of the circuit as a “closed physical system”. Unfortunately, for a
typical AND/OR/NOT circuit, this is not possible. The reason is that the laws of physics governing
microscopic systems (both classical and quantum) are reversible with respect to time, but this is
not true of most gates we would like to physically implement.
• Take for example an AND gate. Suppose its output is |0〉. Can we infer what its inputs were? The
answer is no — they could have been|00〉,|01〉, or |10〉.
• The AND process is not reversible: information sometimes needs to be deleted; “entropy” is lost.
(According to the 2nd Law of Thermodynamics, a physical system consisting of a single AND gate
cannot be “closed”; its operation must dissipate some energy — typically as escaping heat).
• On the other hand, a NOT gate is theoretically “reversible”: its output can be determined from its
input; no information is created or destroyed in switching|0〉to a|1〉or vice versa. Thus, in
principle, it is possible to construct a completely closed physical system implementing a NOT gate,
without the need for energy dissipation.
Contd.
• Definition: A Boolean gate G is said to be reversible if it has the same
number of inputs as outputs, and its mapping from input strings to
output strings (and output to input strings). Thus a NOT gate is
reversible, whereas most other “standard” gates (e.g., AND, OR,
NAND) cannot be reversible since they do not have an equal number
of inputs and outputs.
• For example,
Contd.
• The three examples of reversible gates we have seen so far - NOT,
CNOT, CCNOT - also have the extra property that they are their own
inverse; i.e., applying them twice in succession restores the original
bits. This is a bit of a coincidence, insofar as it is not a property we
insist on for reversible gates. We merely insist that reversible gates
have some inverse gate; the inverse doesn’t have to be the gate itself.
• The CCNOT gate is extremely handy as the following two pictures
show, we can use it to simulate NAND gate (with ancillas & garbage)
as

|1> ancillas
|0> garbage
Contd.
• Note that, in addition to producing the desired NAND outputs, these
conversions also produce extra, unneeded bits (namely, x1 and x2 in
the NAND). This is somewhat inevitable, given that reversible gates
are required to have equally many input and output bits. We call such
unwanted outputs garbage .
•
Contd.
Contd.
• It’s very helpful to think of circuits not just as diagrams but also as
lists of instructions performed on registers", as in the following
description, which is completely equivalent to the diagram in the
Figure given earlier.
•
Randomized computation
• The randomized computation is a bit like the quantum computation - a creative new
twist on classical computation, seemingly realizable in practice and potentially allowing
for big speedups over deterministic computation, but one requiring its investigators to
make a big investment in a new area of math (i.e., probability).
• Indeed, there are some computational tasks which we know how to provably solve
efficiently using randomized computation, but which we don’t know how to provably
solve efficiently using only deterministic computation.
• An example: On input n, generate an n-digit prime number.
• Unlike with quantum computation, however, we believe this is mainly due to our lack of
skill in proving things, rather than an inherent major advantage of randomized
computation.
• For example, we know a deterministic algorithm that we believe efficiently generates
n-digit prime numbers; we just can’t prove its efficiency.
• It is very easy to upgrade the circuit model of computation to a randomized model: we
just introduce a single new gate called the COIN gate, drawn like this:

• It has 0 input and 1 output; the output is a “fair coin flip", viz., |0> with probability 1/2
and |1> with probability 1/2.
Contd.
• Here is an example randomized circuit:
Contd.
Contd.
Quantum Circuit Model of Computation
• A typical quantum circuit with 5 qubit (quantum bit) registers might look like this:

• As usual, it is sufficient to just use CCNOT gates in addition to Hadamard gates (H), but for convenience we also allow NOT and CNOT gates
too.
• Just as in randomized computation, to analyze such a circuit we need to keep track of the joint
state of all 5 qubits as time passes; i.e., as they proceed through the gates. Even though each gate
only affects a small subset of all qubits, nevertheless just as in randomized computation we must
track the state of all qubits at all times in order to keep track of the “correlations", which are
called entanglement in the context of qubits.
Contd.
• Just as in randomized computation, at each time step the state of the 5 qubits is given by an
expression that looks like this:

where the 32 (in general, 2^5) coefficients are possibly negative real numbers.
Since these numbers may be negative, we don’t call them probabilities any
more; we call them amplitudes (and a state like (6) is called a superposition of
|00000> , …, |11111>). Finally, there is a restriction on what amplitudes are
possible; this restriction is that the sum of their squares is always 1:
Contd.
• The rules of analyzing the quantum circuits:
Let us analyze some
basic quantum circuits:
Contd.
• Now when we measure the register, we will always see |0>. It’s kind
of like we “unflipped the coin". The positive/negative cancelation
achieved here is precisely both the power and the mystery of
quantum computation.
Contd.
Postulate 3: Measurement
Contd.
Postulate 4: Multi-qubit systems

Why the tensor product? This is not a proof, but one would expect some way to describe a
composite system. Tensor product works for classical systems (except the restricted definition of
the probability amplitudes makes it so that the result is a simple concatenation). For quantum
systems tensor product captures the essence of superposition, that is if system A is in state | A>
and B in state
|B>, then there should be some way to have a little of A and a little of B. Tensor product exposes
this.
 Entanglements
• Entanglement is a uniquely quantum phenomenon. Entanglement is a
property of a multi-qubit state space (multi-qubit system) and can be
thought of as a resource. To explain entanglement we’ll examine the
creation and destruction of an EPR pair of qubits named after
Einstein, Podolsky, and Rosen.
Contd.
Contd.
• To illustrate why entanglement is so strange, lets consider performing a measurement just prior
to applying the CNot gate. The two measurement operators (for obtaining a |0> or |1> ) are:
Contd.

Hence, after the CNot gate is applied we have only a 1/2 chance of obtaining
|0>.
Contd.
Of particular interest to our discussion, however, is what happens to the state vector of the
system:
Representation of Data

• Quantum computers, which have not been built yet, would be based on the strange
principles of quantum mechanics, in which the smallest particles of light and matter can be in
different places at the same time.
• In a quantum computer, one "qubit" - quantum bit - could be both 0 and 1 at the same time.
So with three qubits of data, a quantum computer could store all eight combinations of 0 and
1 simultaneously. That means a three-qubit quantum computer could calculate eight times
faster than a three-bit digital computer.
• Typical personal computers today calculate 64 bits of data at a time. A quantum computer
with 64 qubits would be 2^64 faster, or about 18 billion billion times faster.
A bit of data is represented by a single atom that is in one of
two states denoted by |0> and |1>. A single bit of this form is
known as a qubit
Representation of Data - Qubits

A physical implementation of a qubit could use the two energy

levels of an atom. An excited state representing |1> and a
ground state representing |0>.

Light pulse of
frequency  for
Excited time interval t
State

Nucleus
Ground
State
Electron
State |0> State |1>
 Representation of Data - Superposition

A single qubit can be forced into a superposition of the two states

denoted by the addition of the state vectors:

|> =  1 |0> +  2 |1>

2 2
Where  1 and 2 are complex numbers and | 1| + |  2 | = 1

A qubit in superposition is in both of the

states |1> and |0 at the same time
Representation of Data - Superposition
Light pulse of
frequency  for time
interval t/2

State |0> State |0> + |1>

Consider a 3 bit qubit register. An equally weighted

superposition of all possible states would be denoted by:
1 1 1
|> = |000> + |001> + . . . + |111>
√8 √8 √8
 Data Retrieval

 In general, an n qubit register can represent the numbers 0

through (2^n) -1 simultaneously.

Sound too good to be true?…It is!

 If we attempt to retrieve the values represented within a
superposition, the superposition randomly collapses to
represent just one of the original values.

In our equation: |> = 1 |0> + 2 |1> ,  represents the

1
probability of the superposition 2 1 to |0>. The ’s
collapsing
are called probability amplitudes. In a balanced
superposition,
n  = 1/√2 n where n is the number of qubits.