# CMSC 39600: PCPs, codes and inapproximability Oct 2, 2007

Lecture 3: Linearity Testing
Lecturer: Prahladh Harsha Scribe: Joshua A. Grochow
In today’s lecture we’ll go over some preliminaries to the coding theory we will need
throughout the course. In particular, we will go over the Walsh-Hadamard code and the
Blum-Luby-Rubinfeld linearity test. In the following lecture we will use these tools to show
that NP ⊆ PCP[poly, O(1)] (recall that our goal is ultimately to bring the polynomial
amount of randomness down to a logarithmic amount). A common feature throughout the
course is that we will use codes to construct PCPs and vice-versa, construct useful codes
from PCPs.
3.1 Coding Theory – Preliminaries
Coding theory is the study of how to communicate reliably over a noisy channel. The most
common setting is as follows: A message m is put through an encoder E, yielding a value
E(m), also called the codeword (typically much longer than m). The codeword E(m) is
then sent through the noisy channel and arrives at the other end with some noise introduced
η by the channel as E(m) + η. The decoder D then takes E(m) + η as input, and ideally
outputs m as long as the noise is not too much. (Note that it is expected of the decoder
that when applied to E(m) it outputs m.)
Formally, a code is speciﬁed by an encoding function C : ¦0, 1¦
k
→ ¦0, 1¦
n
; the outputs
of C are called codewords. The rate of the code C is k/n, i.e. the ratio of input bits to
codeword bits. Heuristically, this is the amount of information of the input message per bit
of the codeword.
An important notion in coding theory is that of the distance between two codewords;
here (and typically) we use either the Hamming distance ∆(x, y) = #¦i[x
i
= y
i
¦ or the
normalized (Hamming) distance δ(x, y) = ∆(x, y)/n (where n = [x[ = [y[).
The distance d of a code C is the minimum distance between two distinct codewords,
i.e., min
x=y
¦∆(C(x), C(y))¦. For a code of distance d, if the number of bits ﬂipped is
strictly less than d/2 (i.e., weight(η) ≤ d/2), then E(m) +η can be uniquely decoded to m.
A code C : ¦0, 1¦
k
→ ¦0, 1¦
n
with distance d is called an (n, k, d)
2
-code, where n is the size
of the codewords, k the size of the input, d the distance of the code, and 2 the size of the
alphabet ¦0, 1¦.
A code C is called linear if for all x and y, if x and y are codewords then so is x + y
(where addition is performed bitwise modulo 2 – i.e. XOR)
1
. To indicate that a (n, k, d)
2
-
code is also linear, we use the notation [n, k, d]
2
-code (with square brackets). The word
“code,” is commonly used to refer to both the encoding function C (as stated above), as
well as the set of all codewords Im(C).
Typically, the most signiﬁcant trade-oﬀ in coding theory is that between the rate and
distance of a code. For example, given a particular rate, we might ask what is the best
1
Note that if we work over a larger alphabet than binary – e.g. over a larger ﬁnite ﬁeld, we require the
additional constraint that x is a codeword then so are all scalar multiples of x (i.e., αx)
3-1
distance that can be achieved.
3.1.1 Algorithmic Questions and Sublinear Time Algorithms
There are three main algorithmic questions that arise in coding theory:
1. Complexity of encoding;
2. Error detection: given r ∈ ¦0, 1¦
n
, decide if r is a codeword; and
3. Error correction: given r ∈ ¦0, 1¦
n
, ﬁnd x ∈ ¦0, 1¦
k
minimizing ∆(r, C(x)).
In this course, we will interested in these questions in the context of sublinear time
algorithms. We need to be speciﬁc what we mean by sublinear time. Note that a sublinear
time algorithm A to compute a function f ¦0, 1¦
k
→ ¦0, 1¦
n
doesn’t have enough time to
read it’s input or write it’s output. We get over this by accessing the input and writing the
output by oracle access. That is, an oracle machine A is a sublinear time algorithm for f if
A
x
(j) = f(x)
j
where f(x)
j
is the j-th bit of f(x). Note that j need only be log n bits, so
can be read entirely in sublinear time. More formally,
• The input is represented implicitly by an oracle. Whenever the sublinear time algo-
rithm wants to access the j
th
bit of the input string x (for some j ∈ [k]), it queries
the input x-oracle for the j
th
bit and obtains x
j
.
Figure 1: Sublinear Time Algorithms
• The output is not explicitly written by the algorithm, instead it is only implicitly
given by the algorithm. Formally, on being queried for index i of the output string
f(x) (for some i ∈ [n]), the algorithm outputs the bit f(x)
i
. Thus, the algorithm itself
behaves as an oracle for the string f(x), which in turn has oracle access to the input
oracle x.
• Since the algorithm does not read the entire input x, we cannot expect it compute
the output f(x) exactly. We instead relax our guarantee on the output as follows:
On input x ∈ ¦0, 1¦
k
, the algorithm must compute f(x

) exactly for some x

∈ ¦0, 1¦
k
that is -close to the actual input x. In other words, the algorithm computes functions
on some approximation to the input instead of the input itself.
3-2
Property testing, for those familiar with it, is typically done in this framework. Figure 1
gives a pictorial description of a sublinear time algorithm with the above mentioned relax-
ations.
Now we’ll consider the algorithmic questions of coding theory in the sublinear time
framework. Let’s consider the questions above one by one:
1. For a code to have good error-correcting properties, most bits of the codeword needs
to depend on most message-bits. Taking this into account, it does not seem reasonable
to expect a “good” code to have sublinear time encoding. (However, there has been
some recent work in the area of locally encodable codes by relaxing some of the error-
correction properties of the code).
2. Error detection: in this context, error detection is known as local testing. That is,
given r ∈ ¦0, 1¦
n
, test if either r is a codeword or r is far from every codeword (similar
to the gap problems we saw earlier, and also to property testing).
3. Error correction is known as local decoding in this context. Given j, the decoder
queries some bits of a noisy codeword C(x) + η and outputs the j-th bit of x. more
formally, we say a code is locally decodable if there exists a local decoder Dec such
that for all x and r where ∆(r, C(x)) < , the decoder satisﬁes
∀i ∈ ¦1, . . . , k¦, Pr[Dec
r
(i) = x
i
] ≥
3
4
.
If there is such a decoder for a given code C and it makes fewer than q queries, then
we say C is (q, )-locally decodable.
Obviously, sublinear time algorithms in general, and local decoding and local testing
algorithms in particular, will be randomized algorithms.
Note that local decoding is only required to work when the input is suﬃciently close to a
codeword, whereas local testability determines whether a given string is close to a codeword
or far from any codeword. Thus the local decodability of a code says nothing about its local
testability.
3.2 The Walsh-Hadamard Code and Linearity Testing
Now onto our ﬁrst code, the Walsh-Hadamard code. This will be the main tool in proving
that NP has exponential size PCPs, i.e. NP ⊆ PCP[poly, O(1)]. There are two dual views
of the Walsh-Hadamard code: on the one hand, WH(x) is the evaluation of every linear
function at x; on the other hand, it consists of the dot product of x with every ∈ ¦0, 1¦
n
.
A function f ¦0, 1¦
k
→ ¦0, 1¦ is linear if ∀x, y, f(x + y) = f(x) + f(y).
For example, for any a ∈ ¦0, 1¦
k
, the function
a
(x) =

a
i
x
i
mod 2 (i.e. the dot
product of a and x as vectors over GF(2)) is a linear function. In fact, L
k
= ¦
a
[a ∈ ¦0, 1¦
k
¦
is the set of all linear boolean functions. (Proof: observe that the space of linear functions
is a vector space, and has the same dimension as L
k
.)
The Walsh-Hadamard code WH : ¦0, 1¦
k
→ ¦0, 1¦
2
k
is then deﬁned as WH(x) =
x
,
i.e. WH(x) is the truth table of
x
. More formally, for any a ∈ ¦0, 1¦
k
, the a-th bit of the
a
=
x
(a).
3-3
Since any two distinct linear functions disagree on exactly half the set of inputs (i.e,
Pr
a
[
x
(a) =
y
(a)] = 1/2, for x = y), the fractional distance of the Walsh-Hadamard codes
is 1/2. Thus the Walsh-Hadamard code is a [k, 2
k
, 2
k−1
]-code. It has very good distance,
but poor rate.
It is useful to note that there are two dual views of the Walsh-Hadamard code based on
the fact that WH(x)
a
=
x
(a) =
a
(x). Thus, WH(x) is both the evaluation of the linear
function
x
at every point as well as the evaluation of every linear function at the point x.
Note that the WH code has the special property that the input bits x
i
are in fact a subset
of the codeword bits, since x
i
=
x
(e
i
). Codes with this property are called systematic codes.
3.2.1 Local Decodability of the Walsh-Hadamard Code
Decoding the Walsh-Hadamard code is very simple. Given a garbled codeword f : ¦0, 1¦
k

¦0, 1¦, which is δ-close to some Walsh-Hadamard codeword, the decoder Dec works as
Dec
f
: “On input z,
1. Choose r ∈
R
¦0, 1¦
k
2. Query f(z + r) and f(r)
3. Output f(z + r) −f(r)

Claim 3.1. If f : ¦0, 1¦
k
→ ¦0, 1¦ is δ-close to WH(x), then,
Pr[Dec
f
(z) =
x
(z)] ≥ 1 −2δ, ∀x ∈ ¦0, 1¦
k
.
Proof. Since f is δ-close to
x
, we have that for a random r, the probability that f(z +r) =

x
(z + r) is at least 1 − δ, and so is the probability that f(r) =
x
(r). If both of these
conditions hold, then Dec
f
(z) =
x
(z), by the linearity of
x
. Thus Pr
r
[Dec
f
(z) =
x
(z)] ≥
1 −2δ.
Since the decoder only makes two queries, the Walsh-Hadamard code is 2-locally decod-
able.
3.2.2 Local Testability of the Walsh-Hadamard Code: Linearity Testing
To locally test the WH code, we wish to test whether a given truth table is the truth table
of a linear function. The problem of local testing of the WH code is more commonly called
linearity testing. Formally, given f ¦0, 1¦
k
→ ¦0, 1¦ we want to test whether it is a linear
function (WH codeword) or far from linear.
The test is, as in the WH decoder, quite simple: pick y and z uniformly at random from
¦0, 1¦
k
and check that f(z) + f(y) = f(z + y). This test was proposed and ﬁrst analyzed
by Blum, Luby and Rubinfeld [BLR93].
BLR-Test
f
: “ 1. Choose y, z ∈
R
¦0, 1¦
k
independently
2. Query f(y), f(z), and f(y + z)
3. Accept if f(y) + f(z) = f(y + z).

3-4
Obviously if f is linear, this test will pass with probability 1. The question is, can there
be function that are far from linear but still pass this test with high probability? No, as
shown in the following
Theorem 3.2. If f is δ-far from linear, then
Pr[BLR-Test
f
accepts ] ≤ 1 −δ.
The above theorem is tight since a random function is 1/2-far from linear, and passes
the BLR-Test with probability exactly 1/2.
The original proof of Blum, Luby and Rubinfeld [BLR93] is a combinatorial proof of
a weaker version of the above theorem, but we will give an algebraic proof, as similar
techniques will arise later in the course. This algebraic proof is due to Bellare, Coppersmith,
+
96]. Before proceeding to the proof, we will ﬁrst need to
equip ourselves with some basics of Boolean Fourier analysis.
3.2.3 Fourier Analysis
First, rather than working over ¦0, 1¦ as the output of a linear function, it will be convenient
to treat the output space as ¦+1, −1¦ (the roots of unity) under multiplication. Thus the
Boolean 0 corresponds to 1 in this setting, and the Boolean 1 corresponds to -1. Linearity
now takes the form: f ¦0, 1¦
n
→ ¦+1, −1¦ is linear if f(x + y) = f(x)f(y).
Consider the family of functions T = ¦f : ¦0, 1¦
k
→ R¦ and equip T with an addition
(f + g)(x) = f(x) + g(x). It is clear that T is a vector space over the reals. Furthermore,
the characteristic functions ¦δ
a
[a ∈ ¦0, 1¦
k
¦ are a basis, where δ
a
(x) = 1 if x = a and 0
otherwise. Thus T has dimension 2
k
.
We will now show that the linear functions of the form χ
a
(x) = (−1)
a(x)
= (−1)
P
a
i
x
i
( mod 2)
also form a basis for T. For this, it is fruitful to deﬁne an inner product on T as follows:
'f, g` = Exp
x∈{0,1}
k [f(x)g(x)] =
1
2
k

x∈{0,1}
k
f(x)g(x).
(It is an easy exercise to check that this is in fact an inner product.) Note that there are
k
= dimT functions of this form, so all we need to do is show that they are linearly
independent.
We begin by examining a few basic properties of the functions χ
a
. First, note that
Property 1. χ
a
(x + y) = χ
a
(x)χ
a
(y)
i.e. χ
a
is linear, as mentioned previously. Second,
Property 2. χ
a+b
(x) = χ
a−b
(x) = χ
a
(x)χ
b
(x)
i.e. χ
a
(x) is also linear in a; this should come as no surprise, because of the duality of

a
(x) as both a linear function of a and of x.
The ﬁrst property we will need that is not entirely obvious is that
Property 3.
Exp
x

a
(x)] =

1 if a = 0
0 otherwise
3-5
Proof. If a = 0, this clearly holds. If a = 0, then by permuting the indices we may assume
that a
1
= 0 without loss of generality. Then we have
2
k
Exp
x

a
(x)] =

x
(−1)
P
a
i
x
i
=

x:x
1
=1
(−1)
P
a
i
x
i
+

x:x
1
=0
(−1)
P
a
i
x
i
Then, since (−1)
a(0y)
= −(−1)
a(1y)
, these two sums exactly cancel out.
We then have,
Property 4.

a
, χ
b
` =

1 if a = b
0 otherwise
.
This follows from the above via 'χ
a
, χ
b
` = E
x

a
(x)χ
b
(x)] = E
x

a−b
(x)], and then
applying the above fact.
Thus, the χ
a
form not only a basis for T, but also an orthonormal basis for T. Since
the χ
a
form an orthonormal basis, for any f ∈ T we may write f =

a
ˆ
f
a
χ
a
for 'f, χ
a
` =
ˆ
f
a
∈ R. These
ˆ
f
a
are called the Fourier coeﬃcients of f.
Observer that if the normalized distance δ(f, χ
a
) = , then
ˆ
f
a
= 1(1−)+(−1) = 1−2,
so the Fourier coeﬃcients capture the normalized distance from linear functions.
Now we come to one of the most basic useful facts in Fourier analysis:
Property 5 (Parseval’s identity). 'f, f` =

ˆ
f
2
a
Proof. Writing f in terms of the basis χ
a
, we get:
'f, f` = '

a
ˆ
f
a
χ
a
,

b
ˆ
f
b
χ
b
`
=

a,b
ˆ
f
a
ˆ
f
b

a
, χ
b
` (by linearity of ', `)
=

a
ˆ
f
2
a
where the last line follows from the previous because the χ
a
form an orthonormal basis.
Corollary 3.3. In particular, if f is a Boolean function, i.e. ±1-valued, then

ˆ
f
2
a
= 1.
3.2.4 Proof of Soundness of BLR-Test
Finally, we come to the proof of the soundness of the Blum-Luby-Rubinfeld linearity test:
Proof of Theorem 3.2. Suppose f is δ-far from any linear function. Note that we can rewrite
the linearity condition f(x)f(y) = f(x + y) as f(x)f(y)f(x + y) = 1, since f is ±1-valued.
Then
Pr
x,y
[BLR-Test accepts f] = Pr
x,y
[f(x)f(y)f(x + y) = 1]
3-6
Note that for any random variable Z with values in ¦+1, −1¦, Pr(Z = 1) = Exp[
1+Z
2
],
since if Z = 1, then (1 + Z)/2 = 1 and if Z = −1, then (1 + Z)/2 = 0, so (1 + Z)/2 is an
indicator variable for the event Z = 1. Thus we have
Pr
x,y
[BLR-Test accepts f] = Exp
x,y

1 + f(x)f(y)f(x + y)
2

=
1
2
+
1
2
Exp
x,y
[f(x)f(y)f(x + y)]
Now, writing out f in terms of its Fourier coeﬃcients, we get
Pr
x,y
[BLR-Test accepts f] =
1
2
+
1
2
Exp
x,y

a,b,c
ˆ
f
a
ˆ
f
b
ˆ
f
c
χ
a
(x)χ
b
(y)χ
c
(x + y)

=
1
2
+
1
2
Exp
x,y

a,b,c
ˆ
f
a
ˆ
f
b
ˆ
f
c
χ
a
(x)χ
b
(y)χ
c
(x)χ
c
(y)

Then, we apply linearity of expectation and the fact that x and y are independent to get:
Pr
x,y
[BLR accepts f] =
1
2
+
1
2

a,b,c
ˆ
f
a
ˆ
f
b
ˆ
f
c
Exp
x

a
(x)χ
c
(x)] Exp
y

b
(y)χ
c
(y)]
=
1
2
+
1
2

a,b,c
ˆ
f
3
a

1
2
+
1
2
max
a
(
ˆ
f
a
)

a
ˆ
f
2
a
=
1
2
+
1
2
max
a
(
ˆ
f
a
) (by Parseval’s identity)
= 1 −δ
where the last line follows from the fact that f is δ-far from linear, so its largest Fourier
coeﬃcient can be at most 1 −2δ, as noted previously.
References
[BCH
+
96] Mihir Bellare, Don Coppersmith, Johan H˚astad, Marcos A. Kiwi, and
Madhu Sudan. Linearity testing in characteristic two. IEEE Transactions on
Information Theory, 42(6):1781–1795, November 1996. (Preliminary version in
36th FOCS, 1995). doi:10.1109/18.556674.
[BLR93] Manuel Blum, Michael Luby, and Ronitt Rubinfeld. Self-
testing/correcting with applications to numerical problems. J. Computer and
System Sciences, 47(3):549–595, December 1993. (Preliminary Version in 22nd
STOC, 1990). doi:10.1016/0022-0000(93)90044-W.
3-7

Figure 1: Sublinear Time Algorithms • The output is not explicitly written by the algorithm. we cannot expect it compute the output f (x) exactly. 3. ﬁnd x ∈ {0. 3-2 . In this course. 1}k → {0. an oracle machine A is a sublinear time algorithm for f if Ax (j) = f (x)j where f (x)j is the j-th bit of f (x). 1}k minimizing ∆(r. so can be read entirely in sublinear time. We instead relax our guarantee on the output as follows: On input x ∈ {0. which in turn has oracle access to the input oracle x. Note that j need only be log n bits. instead it is only implicitly given by the algorithm. C(x)). • Since the algorithm does not read the entire input x. it queries the input x-oracle for the j th bit and obtains xj .1 Algorithmic Questions and Sublinear Time Algorithms There are three main algorithmic questions that arise in coding theory: 1. on being queried for index i of the output string f (x) (for some i ∈ [n]). 1}n . Note that a sublinear time algorithm A to compute a function f {0. We need to be speciﬁc what we mean by sublinear time. we will interested in these questions in the context of sublinear time algorithms. 2. 1}k that is -close to the actual input x. Error correction: given r ∈ {0. Error detection: given r ∈ {0. the algorithm must compute f (x ) exactly for some x ∈ {0.distance that can be achieved. 1}n . 1}k . That is. Formally. Whenever the sublinear time algorithm wants to access the j th bit of the input string x (for some j ∈ [k]). decide if r is a codeword. the algorithm computes functions on some approximation to the input instead of the input itself. Complexity of encoding. More formally.1. the algorithm itself behaves as an oracle for the string f (x). 1}n doesn’t have enough time to read it’s input or write it’s output. the algorithm outputs the bit f (x)i . We get over this by accessing the input and writing the output by oracle access. and 3. • The input is represented implicitly by an oracle. In other words. Thus.

and has the same dimension as Lk . 4 If there is such a decoder for a given code C and it makes fewer than q queries. it does not seem reasonable to expect a “good” code to have sublinear time encoding. . i. will be randomized algorithms. More formally. For example. and local decoding and local testing algorithms in particular. (Proof: observe that the space of linear functions is a vector space. we say a code is locally decodable if there exists a local decoder Dec such that for all x and r where ∆(r. 1}k → {0. This will be the main tool in proving that N P has exponential size PCPs. the a-th bit of the Walsh-Hadamard codeword W H(x) is W H(x)a = x (a). 2. 1} is linear if ∀x. 3-3 . for those familiar with it. (However. In fact. test if either r is a codeword or r is far from every codeword (similar to the gap problems we saw earlier.e. Given j. the dot product of a and x as vectors over GF (2)) is a linear function. 3.2 The Walsh-Hadamard Code and Linearity Testing Now onto our ﬁrst code.Property testing. Lk = { a |a ∈ {0. 1}k → {0. the decoder satisﬁes 3 ∀i ∈ {1. . on the other hand. there has been some recent work in the area of locally encodable codes by relaxing some of the errorcorrection properties of the code). .e. W H(x) is the evaluation of every linear function at x. for any a ∈ {0. the Walsh-Hadamard code. Note that local decoding is only required to work when the input is suﬃciently close to a codeword. Let’s consider the questions above one by one: 1. 1}k .) k The Walsh-Hadamard code W H : {0. Now we’ll consider the algorithmic questions of coding theory in the sublinear time framework. C(x)) < . k}. f (x + y) = f (x) + f (y). That is. is typically done in this framework. A function f {0. more formally. Taking this into account. Pr[Decr (i) = xi ] ≥ . Obviously. 1}k } is the set of all linear boolean functions. sublinear time algorithms in general. it consists of the dot product of x with every ∈ {0. whereas local testability determines whether a given string is close to a codeword or far from any codeword. 1}n . the function a (x) = ai xi mod 2 (i. . N P ⊆ P CP [poly. For a code to have good error-correcting properties. then we say C is (q. most bits of the codeword needs to depend on most message-bits. the decoder queries some bits of a noisy codeword C(x) + η and outputs the j-th bit of x.e. for any a ∈ {0. O(1)]. Error correction is known as local decoding in this context. y. Thus the local decodability of a code says nothing about its local testability. and also to property testing). Figure 1 gives a pictorial description of a sublinear time algorithm with the above mentioned relaxations. 3. given r ∈ {0. There are two dual views of the Walsh-Hadamard code: on the one hand. 1}k . )-locally decodable. 1}n . 1}2 is then deﬁned as W H(x) = x . W H(x) is the truth table of x . error detection is known as local testing. Error detection: in this context. i.

It is useful to note that there are two dual views of the Walsh-Hadamard code based on the fact that W H(x)a = x (a) = a (x). by the linearity of x . given f {0. 2k−1 ]-code. quite simple: pick y and z uniformly at random from {0. Choose y. Formally. but poor rate. Query f (z + r) and f (r) 3. 1}k and check that f (z) + f (y) = f (z + y). The problem of local testing of the WH code is more commonly called linearity testing. we have that for a random r. 3. Since the decoder only makes two queries. we wish to test whether a given truth table is the truth table of a linear function. W H(x) is both the evaluation of the linear function x at every point as well as the evaluation of every linear function at the point x. 1}k → {0. Choose r ∈R {0. If f : {0. 2k . Thus the Walsh-Hadamard code is a [k. 1}k independently 2. the fractional distance of the Walsh-Hadamard codes is 1/2. as in the WH decoder. 1}. The test is. If both of these conditions hold.Since any two distinct linear functions disagree on exactly half the set of inputs (i. Thus Prr [Decf (z) = x (z)] ≥ 1 − 2δ. Codes with this property are called systematic codes.2 Local Testability of the Walsh-Hadamard Code: Linearity Testing To locally test the WH code. Output f (z + r) − f (r) Claim 3. since xi = x (ei ). Since f is δ-close to x . Accept if f (y) + f (z) = f (y + z). and f (y + z) 3.2. Proof. the probability that f (z + r) = x (z + r) is at least 1 − δ. 3-4 . which is δ-close to some Walsh-Hadamard codeword.1 Local Decodability of the Walsh-Hadamard Code Decoding the Walsh-Hadamard code is very simple. Pra [ x (a) = y (a)] = 1/2. Thus. 1}k 2. Given a garbled codeword f : {0. the Walsh-Hadamard code is 2-locally decodable. 1} we want to test whether it is a linear function (WH codeword) or far from linear. 1}k . 1} is δ-close to W H(x). ∀x ∈ {0. This test was proposed and ﬁrst analyzed by Blum. and so is the probability that f (r) = x (r). for x = y).e.1. BLR-Testf : “ 1. Luby and Rubinfeld [BLR93]. Query f (y). 3. It has very good distance. Note that the WH code has the special property that the input bits xi are in fact a subset of the codeword bits.2. then Decf (z) = x (z). z ∈R {0. 1}k → {0. the decoder Dec works as follows: (The decoder Dec has oracle access to the function f ) Decf : “On input z. f (z). Pr[Decf (z) = x (z)] ≥ 1 − 2δ. 1. 1}k → {0. then.

can there be function that are far from linear but still pass this test with high probability? No. then Pr[BLR-Testf accepts ] ≤ 1 − δ. as shown in the following Theorem 3.2.e. x∈{0. χa is linear. this should come as no surprise. Linearity now takes the form: f {0. 3. Property 2.2. and passes the BLR-Test with probability exactly 1/2. If f is δ-far from linear. −1} (the roots of unity) under multiplication.1}k [f (x)g(x)] = 1 · 2k f (x)g(x). 1} as the output of a linear function. This algebraic proof is due to Bellare. and the Boolean 1 corresponds to -1. Coppersmith. but we will give an algebraic proof. We begin by examining a few basic properties of the functions χa . 1}n → {+1.1}k (It is an easy exercise to check that this is in fact an inner product. Furthermore. it will be convenient to treat the output space as {+1. First. The original proof of Blum. For this. this test will pass with probability 1. χa (x) is also linear in a. as mentioned previously. 1}k → R} and equip F with an addition (f + g)(x) = f (x) + g(x). Consider the family of functions F = {f : {0. rather than working over {0. h˚ astad. as similar techniques will arise later in the course. a The ﬁrst property we will need that is not entirely obvious is that Property 3. because of the duality of (x) as both a linear function of a and of x. Luby and Rubinfeld [BLR93] is a combinatorial proof of a weaker version of the above theorem. It is clear that F is a vector space over the reals. χa+b (x) = χa−b (x) = χa (x)χb (x) i. where δa (x) = 1 if x = a and 0 otherwise.e. The question is.3 Fourier Analysis First. The above theorem is tight since a random function is 1/2-far from linear. Thus F has dimension 2k . χa (x + y) = χa (x)χa (y) i.Obviously if f is linear. so all we need to do is show that they are linearly independent. −1} is linear if f (x + y) = f (x)f (y). the characteristic functions {δa |a ∈ {0.) Note that there are already 2k = dim F functions of this form. g = Expx∈{0. P We will now show that the linear functions of the form χa (x) = (−1) a (x) = (−1) ai xi ( mod 2) also form a basis for F. 1}k } are a basis. we will ﬁrst need to equip ourselves with some basics of Boolean Fourier analysis. Before proceeding to the proof. Second. it is fruitful to deﬁne an inner product on F as follows: f. Expx [χa (x)] = 1 if a = 0 0 otherwise 3-5 . Thus the Boolean 0 corresponds to 1 in this setting. Kiwi and Sudan [BCH+ 96]. note that Property 1.

3.2. since (−1) a (0y) (−1) x:x1 =1 P ai xi + (−1) x:x1 =0 P ai xi = −(−1) a (1y) . we get: f. for any f ∈ F we may write f = a fa χa for f. Note that we can rewrite the linearity condition f (x)f (y) = f (x + y) as f (x)f (y)f (x + y) = 1. otherwise This follows from the above via χa .3. f = a ˆ fa χa . In particular. but also an orthonormal basis for F. these two sums exactly cancel out.b ˆ ˆ fa fb χa .4 Proof of Soundness of BLR-Test Finally. b ˆ fb χb = a.e. so the Fourier coeﬃcients capture the normalized distance from linear functions.y 3-6 . f. χa = ˆ ˆ fa ∈ R. ˆ Observer that if the normalized distance δ(f. i. Corollary 3. χb (by linearity of ·. ±1-valued. then by permuting the indices we may assume that a1 = 0 without loss of generality. We then have. Thus. · ) ˆ2 fa a = where the last line follows from the previous because the χa form an orthonormal basis. Then we have 2k · Expx [χa (x)] = x (−1) P ai xi = Then. Writing f in terms of the basis χa . and then applying the above fact. These fa are called the Fourier coeﬃcients of f . Property 4. since f is ±1-valued. if f is a Boolean function. Now we come to one of the most basic useful facts in Fourier analysis: Property 5 (Parseval’s identity). Suppose f is δ-far from any linear function. we come to the proof of the soundness of the Blum-Luby-Rubinfeld linearity test: Proof of Theorem 3.2. χa . the χa form not only a basis for F. Then Pr [BLR-Test accepts f ] = Pr [f (x)f (y)f (x + y) = 1] x. χa ) = . Since ˆ the χa form an orthonormal basis. χb = 1 0 if a = b . χb = Ex [χa (x)χb (x)] = Ex [χa−b (x)]. f = ˆ2 fa Proof. If a = 0. this clearly holds.y x. If a = 0.Proof. then fa = 1(1− )+(−1) = 1−2 . then ˆ2 fa = 1.

−1}.b.y x. so its largest Fourier coeﬃcient can be at most 1 − 2δ. Marcos A.Note that for any random variable Z with values in {+1. 47(3):549–595. 3-7 . December 1993. 2 since if Z = 1.1109/18. Don Coppersmith. Selftesting/correcting with applications to numerical problems.c 1 1 ˆ + max(fa ) 2 2 a ˆ2 fa a 1 1 ˆ + max(fa ) (by Parseval’s identity) = 2 2 a = 1−δ where the last line follows from the fact that f is δ-far from linear. 1995).c   1 1 ˆ ˆˆ = + Expx. J.c Then.y 2 2 a. Michael Luby. then (1 + Z)/2 = 0. Thus we have Pr [BLR-Test accepts f ] = Expx. (Preliminary version in 36th FOCS. Computer and System Sciences.y  fa fb fc χa (x)χb (y)χc (x)χc (y) 2 2 a. P r(Z = 1) = Exp[ 1+Z ].y [f (x)f (y)f (x + y)] 2 2 Now. so (1 + Z)/2 is an indicator variable for the event Z = 1.y 1 + f (x)f (y)f (x + y) 2 = 1 1 + Expx. November 1996. Johan H˚ astad.y 1 1 + 2 2 1 1 + 2 2 ˆ ˆˆ fa fb fc Expx [χa (x)χc (x)] Expy [χb (y)χc (y)] a. as noted previously. IEEE Transactions on Information Theory. (Preliminary Version in 22nd STOC. writing out f in terms of its Fourier coeﬃcients. References [BCH+ 96] Mihir Bellare.b. 1990).y  fa fb fc χa (x)χb (y)χc (x + y) x. [BLR93] Manuel Blum. doi:10.556674. we apply linearity of expectation and the fact that x and y are independent to get: Pr [BLR accepts f ] = x. 42(6):1781–1795.c = ≤ ˆ3 fa a. Linearity testing in characteristic two.1016/0022-0000(93)90044-W. we get   1 1 ˆ ˆˆ Pr [BLR-Test accepts f ] = + Expx. and Ronitt Rubinfeld.b. and Madhu Sudan. then (1 + Z)/2 = 1 and if Z = −1.b. Kiwi. doi:10.