HAL Id: hal-03219028
https://hal.archives-ouvertes.fr/hal-03219028
Submitted on 6 May 2021
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Point-to-Point Strategic Communication
Mael Le Treust, Tristan Tomala
To cite this version:
Mael Le Treust, Tristan Tomala. Point-to-Point Strategic Communication. IEEE ITW, Oct 2020,
Riva del Garda (virtual), Italy. �hal-03219028�
Point-to-Point Strategic Communication
Maël Le Treust
ETIS UMR 8051, CY Cergy Paris Université, ENSEA, CNRS, 6, avenue du Ponceau,
95014 Cergy-Pontoise CEDEX, France Email: mael.le-treust@ensea.fr
Tristan Tomala
HEC Paris, GREGHEC UMR 2959 1 rue de la Libération, 78351 Jouy-en-Josas CEDEX, France
Email: tomala@hec.fr
Abstract
—We investigate a strategic formulation of the joint source-channel coding problem in which the encoder and the decoder are endowed with distinct distortion functions. We provide the solutions in four different scenarios. First, we assume that the encoder and the decoder cooperate in order to achieve a certain pair of distortion values. Second, we suppose that the encoder commits to a strategy whereas the decoder implements a best response, as in the persuasion game where the encoder is the Stackelberg leader. Third, we consider that the decoder commits to a strategy, as in the mismatched rate-distortion problem or as in the mechanism design framework. Fourth, we study the cheap talk game in which the encoding and the decoding strategies form a Nash equilibrium.
I. I
NTRODUCTIONStrategic communication takes place when an informed sender communicates with a receiver that takes an action, given that the sender and the receiver optimize different metrics. This question was originally formulated in the Game Theory literature were the messages are costless and the communication is unrestricted. Crawford and Sobel [1] inves- tigate the Nash equilibrium of the cheap talk game, whereas Kamenica and Gentzkow [2] introduce the Bayesian persua- sion game in which the sender commits to an information disclosure policy, as the leader of the Stackelberg game. In a previous work [3], we characterize the solution of the Bayesian persuasion game when the communication channel is noisy.
U
nX
nY
nV
nP
Uσ T
Y|Xτ
d
e(u, v) d
d(u, v)
Fig. 1. The source is i.i.d. and the channel is memoryless. The encoder and the decoder have mismatched distortion functionsde(u, v)6=dd(u, v).
The strategic communication problem has attracted attention in Computer Science [4], in Control Theory [5], in Information Theory [6], [7], [8], [9] and is related to the lossy source coding with mismatch distortion functions [10], [11]. Recently, Vora and Kulkarni investigate a strategic communication prob- lem in which the receiver is the Stackelberg leader that must recover the source sequence [12]. The authors introduce the notion of the “information extraction capacity” and formulate an elegant solution in terms of the zero error capacity of “the sender graph” [13].
In this paper, we compare four different solutions for the point-to-point strategic communication problem and we characterize the limit set of ε-Nash equilibrium distortions.
II. S
YSTEMM
ODELWe denote by U , X , Y , V, the finite sets of information source, channel inputs, channel outputs and decoder’s outputs.
Uppercase letters U
n= (U
1, . . . , U
n) ∈ U
nand X
n, Y
n, V
nstand for n-length sequences of random variables with n ∈ N
⋆= N\ {0}, whereas lowercase letters u
n= (u
1, . . . , u
n) ∈ U
nand x
n, y
n, v
n, stand for sequences of realizations. We denote by ∆(X) the set of probability distributions Q
Xover X , i.e. the probability simplex. The support of a distribution Q
Xis denoted by supp Q
X= {x ∈ X, Q(x) > 0}.
We consider an i.i.d. information source and a memoryless channel distributed according to P
U∈ ∆(U ) and T
Y|X: X →
∆(Y), as depicted in Fig. 1.
Definition 1 We define the encoding strategy σ : U
n−→
∆(X
n) and the decoding strategy τ : Y
n−→ ∆(V
n), and we denote by P
σ,τthe distribution defined by
P
σ,τ= Y
nt=1
P
Utσ
Xn|UnY
nt=1
T
Yt|Xtτ
Vn|Yn, (1) where σ
Xn|Un, τ
Vn|Yndenote the distributions of σ, τ.
Definition 2 The encoder and decoder distortion functions d
e: U × V −→ R and d
d: U × V −→ R induce long-run distortion functions d
en(σ, τ ) and d
dn(σ, τ) defined by
d
dn(σ, τ ) = X
un,vn
P
σ,τu
n, v
n·
"
1 n
X
nt=1
d
d(u
t, v
t)
# . (2) III. C
OOPERATIVES
CENARIODefinition 3 The pair (D
e, D
d) is achievable if
∀ε > 0, ∃¯ n ∈ N
⋆, ∀n ≥ n, ¯ ∃(σ, τ ), (3)
|D
e− d
en(σ, τ )| + |D
d− d
dn(σ, τ )| ≤ ε (4) We denote by C the set of achievable pairs (D
e, D
d).
We define the set of distributions e
Q = n
P
UQ
V|Us.t. max
PX
I(X; Y ) − I(U ; V ) ≥ 0 o . (5) Theorem 1 (Cooperative scenario)
C = n
E
Q[d
e(U, V )], E
Q[d
d(U, V )]
Q ∈ Q e o
. (6)
The proof of Theorem 1 follows from Shannon’s separation
result [15, Theorem 3.7] with two distortion functions.
IV. P
ERSUASIONG
AME: E
NCODERC
OMMITMENTIn this section, the encoder chooses first a strategy σ and the decoder selects a best response strategy τ accordingly. This corresponds to the Bayesian persuasion game [2], where the encoder is the Stackelberg leader.
Definition 4 Given n ∈ N
⋆, we define
1. the set of decoder best responses to strategy σ by
BR
d(σ) =argmin
τ
d
dn(σ, τ ), (7) 2. the long-run encoder distortion value by
D
en= inf
σ
max
τ∈BRd(σ)
d
en(σ, τ ). (8) In case BR
d(σ) is not a singleton, we assume that the decoder selects the worst strategy for the encoder distortion max
τ∈BRd(σ)d
en(σ, τ ), so that the solution is robust to the exact specification of the decoding strategy.
We aim at characterizing the asymptotic behavior of D
en. Definition 5 We consider an auxiliary random variable W ∈ W with |W| = min |U| + 1, |V|
and we define
Q = n
P
UQ
W|U, max
PX
I(X ; Y ) − I(U ; W ) ≥ 0 o . (9) Given Q
U W, we define the single-letter decoder best responses
A
dQ
U W=argmin
QV|W
E
QU W QV|Wh d
d(U, V ) i
. (10)
The encoder optimal distortion D
e⋆is given by D
e⋆= inf
QU W∈Q
max
QV|W∈ Ad(QU W)
E
QU W QV|Wh
d
e(U, V ) i
. (11)
Theorem 2 (Encoder commitment, Theorem 3.1 in [3])
∀n ∈ N
⋆, D
ne≥ D
e⋆, (12)
∀ε > 0, ∃¯ n ∈ N
⋆, ∀n ≥ n, ¯ D
ne≤ D
e⋆+ ε. (13) Theorem 2 is a particular case of [9, Theorem III.3] when no side information is available at the decoder. Note that the sequence (D
ne)
n∈N⋆is sub-additive. Indeed, when σ is the concatenation of several encoding strategies, the concatenation of the corresponding optimal decoding strategies still belongs to BR
d(σ). Theorem 2 and Fekete’s lemma show that
D
⋆e= lim
n→+∞
D
ne= inf
n∈N⋆
D
en. (14) Remark 1 The decoder long-run distortion d
dn(σ, τ ) obtained with σ asymptotically optimal for (8) and τ ∈ BR
d(σ), converges to E
QU WQV|W
d
d(U, V )
, where Q
V|W∈ A
dQ
U Wand Q
U Wis a limit of a minimizing sequence of (11).
V. M
ECHANISMD
ESIGN: D
ECODERC
OMMITMENTIn this section, it is the decoder that chooses first a strategy τ and then the encoder selects a strategy σ accordingly.
This corresponds to the mismatched rate-distortion problem in Information Theory [10], [11] and to the mechanism design problem [14] in Game Theory, where the decoder is the Stackelberg leader.
Definition 6 Given n ∈ N
⋆, we define
1. the set of encoder best responses to strategy τ by BR
e(τ ) =argmin
σ
d
en(σ, τ ), (15) 2. the long-run decoder distortion value by
D
nd= inf
τ
max
σ∈BRe(τ)
d
dn(σ, τ ). (16) The value D
ndcorresponds to the best distortion the decoder can obtain for fixed n ∈ N
⋆. In case there are several best responses, we assume the encoder selects the worst strategy σ for the decoder distortion.
We aim at characterizing the asymptotic behaviour of D
dnDefinition 7 Given an auxiliary random variable W ∈ W with |W| = min |U|+1, |V|
with distribution P
W, we define P(P
W) = n
Q
U W∈ ∆(U × W), Q
U= P
U, Q
W= P
W, max
PX
I(X; Y ) − I(U; W ) ≥ 0 o
. (17) Given P
W V, we define the single-letter encoder best responses
A
e(P
W V) = argmin
QU W∈P(PW)
E
QU W PV|Wh d
e(U, V ) i
. (18)
The decoder optimal distortion D
⋆dis given by D
d⋆= inf
PW V Q
max
U W∈ Ae(PW V)E
QU W PV|Wh d
d(U, V ) i
. (19)
In both (11) and (19), it is the Stackelberg leader that selects the marginal distribution P
W, whereas the incentive constraints affect the Stackelberg follower. In both settings, the encoder selects the distribution Q
U W∈ P(P
W) that satisfies the information constraint and the decoder selects P
V|W. Theorem 3 (Decoder commitment)
∀n ∈ N
⋆, D
nd≥ D
⋆d, (20)
∀ε > 0, ∃¯ n ∈ N
⋆, ∀n ≥ ¯ n, D
nd≤ D
⋆d+ ε. (21) The achievability proof of Theorem 3 is provided in App.
B, and relies on similar arguments as in [10, Step 1] and [11, Lemma 4.3]. The converse proof is based on the identification of the auxiliary random variables U = U
T, W = (Y
n, T ), V = V
Twhere T ∈ {1, . . . , n} is uniformly distributed.
The sequence (D
dn)
n∈N⋆is sub-additive, thus Theorem 3 and Fekete’s lemma show that
D
d⋆= lim
n→+∞
D
dn= inf
n∈N⋆
D
nd. (22)
Remark 2 The encoder long-run distortion d
en(σ, τ ) obtained with τ asymptotically optimal for (16) and σ ∈ BR
e(τ), converges to E
QU WPV|W
d
e(U, V )
, where Q
U W∈ A
eP
W Vand P
W Vis a limit of a minimizing sequence of (19).
VI. C
HEAPT
ALKG
AME: N
OC
OMMITMENTDefinition 8 Given ε ≥ 0 and n ∈ N
⋆, an ε-Nash equilibrium is a pair of strategies (σ, τ) such that
σ ∈ BR
eε(τ) and τ ∈ BR
dε(σ) where, (23) BR
εe
(τ) = n
σ, d
en(σ, τ ) ≤ min
˜
σ
d
en(˜ σ, τ ) + ε o
, (24) BR
εd(σ) = n
τ, d
dn(σ, τ ) ≤ min
˜
τ
d
dn(σ, τ) + ˜ ε o
. (25) We denote by NE
nε
the set of distortion pairs (D
εe, D
εd) for which there exists a ε-Nash equilibrium (σ, τ ) such that
D
εe= d
en(σ, τ ) and D
dε= d
dn(σ, τ ). (26) We denote by NE
nthe set NE
nε
with ε = 0.
Definition 9 For ε ≥ 0, we define the set of distributions that are ε-best responses for both encoder and decoder.
D
ε= n
Q
U W V= P
UQ
W|UQ
V|Ws.t.
Q
U W∈ A
εe(Q
W V), Q
V|W∈ A
εd(Q
U W) o , (27) A
εe(Q
W V) = n
Q
U W∈ P (Q
W) s.t. E
QU W QV|Wh d
e(U, V ) i
≤ min
QeU W
∈P(QW)
E
QeU W QV|Wh
d
e(U, V ) i + ε o
, (28)
A
εd(Q
U W) = n
Q
V|Ws.t. E
QU W QV|Wh
d
d(U, V ) i
≤ min
PeV|W
E
QU W e PV|Wh d
d(U, V ) i + ε o
. (29)
Then, we define N
ε= n
E
Q[d
e(U, V )], E
Q[d
d(U, V )]
, Q ∈ D
εo . (30) We denote by N (resp. D) the set N
ε(resp. D
ε) with ε = 0.
Theorem 4 (Nash equilibrium distortions)
∀ε ≥ 0, ∀n ∈ N, NE
nε
⊂ N
ε, (31)
ε→0
lim lim
n→+∞
NE
nε
= N . (32)
Theorem 4 is a consequence of the arguments in the achievability proofs of Theorems 2 and 3, which show that Shannon’s encoding and decoding schemes based a distribu- tion P
UQ
W|UQ
V|W∈ D
ε, form an ε-Nash equilibrium.
Conjecture 1
n→+∞
lim lim
ε→0
NE
nε
= N . (33)
Proposition 1 We compare the solutions (6), (11), (19), (30).
min
QV|U∈eQ
E
d
e(U, V )
≤ D
e⋆≤ inf
QU W V∈D
E
d
e(U, V ) , (34) min
QV|U∈eQ
E
d
d(U, V )
≤ D
d⋆≤ inf
QU W V∈D
E
d
d(U, V ) . (35)
A
PPENDIXA P
RELIMINARYR
ESULTSDefinition 10 Given P
U W∈ ∆(U × W), tolerance δ > 0, let B
δ(P
U W) = n
Q
U Ws.t. ||Q
U W− P
U W||
1≤ δ o
. (36) We define the set of typical sequences by
T
δ(P
U W) = n
(u
n, w
n) s.t. Q
nU W∈ B
δ(P
U W) o
, (37) where Q
nU Wdenotes the empirical distribution of (u
n, w
n).
Definition 11 We consider two distributions P
U∈ ∆(U ), P
W∈ ∆(W), a rate parameter R ≥ 0 and a tolerance δ ≥ 0.
We define the sets
Q
−δ( R ) = n
Q
U W∈ ∆(U × W ) s.t. ||Q
U− P
U||
1≤ δ,
||Q
W− P
W||
1≤ δ and I(U ; W ) ≤ R o
, (38) Q
+δ( R ) = n
Q
U W∈ ∆(U × W ) s.t. ||Q
U− P
U||
1≤ δ,
||Q
W− P
W||
1≤ δ and I(U ; W ) ≥ R o
. (39) We use the notation Q
−0( R ) and Q
+0( R ) when δ = 0.
Lemma 1 (see Step 1 in [10] and Lemma 4.3 in [11]) We consider two distributions P
U∈ ∆(U ) and P
W∈ ∆(W), a rate R ≥ 0, a small parameter η > 0 and n ∈ N
⋆.
• We generate a sequence U
naccording to P
U⊗n.
• Independently, we generate a family of sequences W
n(m)
m∈{1,...,2nR}
according to P
W⊗n.
There exists δ, for all ¯ δ < δ ¯ and for all ε > 0, there exists n, ¯ for all n ≥ ¯ n,
Pr
∃m ∈ {1, . . . , 2
nR}, Q
nm∈ Q
+δ( R + η)
≤ ε,
where Q
nmdenotes the empirical distribution of (U
n, W
n(m)).
The provide the proof of Lemma 1 in App. C.
Lemma 2 (Covering lemma, see Lemma 3.3 in [15]) We consider a distribution P
U W∈ ∆(U × W), a rate parameter R = I(U ; W ) + η with η > 0, n ∈ N.
• We generate a sequence U
naccording to P
U⊗n.
• Independently, we generate a family of sequences W
n(m)
m∈{1,...,2nR}
according to P
W⊗n.
There exists ¯ δ > 0, for all δ < δ ¯ and for all ε > 0, there exists
¯
n, such that for all n ≥ n, ¯ Pr
∃m ∈ {1, . . . , 2
nR}, ||Q
nm− P
U W||
1≤ δ
≥ 1 − ε.
Definition 12 For P
U∈ ∆(U ), P
W V∈ ∆(W × V ), δ > 0, R ≥ 0, and D ≥ 0 we define
Q
δ( R , D ) = n
Q
U W∈ ∆(U × W ) s.t. ||Q
U− P
U||
1≤ δ,
||Q
W− P
W||
1≤ δ, I(U ; W ) ≤ R , E
QU W PV|Wh d
e(U, V ) i
≤ D o
.
(40)
We have Q
δ( R , D ) = Q
−δ( R ) ∩ Q
◦δ( D ) with Q
◦δ( D ) = n
Q
U W∈ ∆(U × W) s.t. ||Q
U− P
U||
1≤ δ,
||Q
W− P
W||
1≤ δ and E
QU W PV|Wh d
e(U, V ) i
≤ D o . (41) A
PPENDIXB
A
CHIEVABILITYP
ROOF OFT
HEOREM3
If the channel capacity is equal to zero, then a trivial coding scheme satisfies (21). From now on, we assume that the channel capacity is strictly positive. Therefore, for all ε
0> 0 there exists η
0> 0 and a distribution P
W Vsuch that
D
d⋆− max
QU W∈Aηe0(PW V)
E
QU W PV|Wh d
d(U, V ) i ≤ ε
0, (42)
where
A
ηe0(P
W V) = argmin
QU W∈Pη0(PW)
E
QU W PV|Wh d
e(U, V ) i
, (43)
P
η0(P
W) = n
Q
U W∈ ∆(U × W) s.t. Q
U= P
U, Q
W= P
Wand max
PX
I(X; Y ) − I(U; W ) ≥ 2η
0o . (44) We use the notation Q
U Wto refer to the distribution that achieves the maximum in (42), and without loss of generality, we assume that I(U ; W ) = max
PXI(X ; Y ) − 2η
0. We introduce the rate parameter R = I(U ; W ) + η
0and the tolerance of the typical sequences δ > 0. We consider that the decoder implements Shannon’s channel decoding and lossy source decoding, see [15, Sec. 3.1 and 3.6], that we denote by τ
⋆. We denote by M and m the indexes selected by the encoder, whereas M ˆ and m ˆ refer to the indexes selected by the decoder.
• The random codebooks (W
n(m), X
n(m))
m∈{1,...,2nR}are drawn independently according to P
W⊗nand P
X⊗n, where P
Xmaximizes the channel capacity.
• The decoder observes the sequence of channel output Y
n∈ Y
nand returns the unique index m ˆ such that the sequences Y
n, X
n( ˆ m)
∈ T
δ(P
XT
Y|X) are jointly typical. Otherwise it returns the index 1.
• Then the decoder returns the sequence W
n( ˆ m) corre- sponding to m ˆ and draws V
ni.i.d. according to P
V|W. Standard channel coding arguments ensures that
∃ δ ¯
1, ∀δ < δ ¯
1, ∀ε
1, ∃¯ n
1, ∀n ≥ n ¯
1, Pr( ˆ M 6= M ) ≤ ε
1. (45) Since the encoder is strategic, it selects a best response σ ∈ BR
e(τ
⋆) that, for a given u
n, returns x
nin order to minimize
X
yn ,vn ˆ m
T
⊗n(y
n|x
n)Pr( ˆ m|y
n)P
⊗n(v
n|w
n( ˆ m)) 1 n
X
nt=1
d
e(u
t, v
t)
= X
mˆ
Pr( ˆ m|x
n) · X
u,w
Q
nmˆ(u, w) X
v
P(v|w)d
e(u, v), (46) where Q
nmˆ∈ ∆(U × W) denotes the empirical distribution of (u
n, w
n( ˆ m)). We denote by x
n⋆the sequence that minimizes (46) and we denote by
Q
xn= X
mˆ
Pr( ˆ m|x
n) · Q
nmˆ∈ ∆(U × W), (47)
the average empirical distribution induced by the input se- quence x
n. By Lemma 1, for all η
2> 0, there exists δ ¯
2, for all δ < δ ¯
2and for all ε
2> 0, there exists ¯ n
2, for all n ≥ n ¯
2,
Pr
Q
Xn⋆∈ / Q
−δ( R + η
2)
≤ Pr
Q
Xn⋆∈ Q
+δ( R + η
2)
(48) + Pr
||Q
XUn⋆− P
U||
1+ ||Q
XWn⋆− P
W||
1> δ
(49)
≤Pr
∃x
n∈ X
n, Q
xn∈ Q
+δ( R + η
2)
+ ε
2(50)
≤Pr
∃m ∈ {1, . . . , 2
nR}, Q
nm∈ Q
+δ( R + η
2)
+ ε
2(51)
≤2ε
2. (52)
On the other hand, we assume that the encoder implements Shannon’s coding scheme σ
c, by selecting the unique m such that (U
n, W
n(m)) ∈ T
δ(Q
U W), and 1 otherwise. By Lemma 2, there exists δ ¯
3> 0, for all δ < δ ¯
3and for all ε
3> 0, there exists n ¯
3, such that for all n ≥ ¯ n
3,
Pr
∀m ∈ {1, . . . , 2
nR}, ||Q
nm− Q
U W||
1> δ
≤ ε
3. (53) The bounds given in (45), (53) imply
1 − ε
1− ε
3≤ Pr
Q
Xn(m)∈ Q
◦δ( D + µ)
(54)
≤ Pr
Q
Xn⋆∈ Q
◦δ( D + µ)
, (55)
with D = min
QU W∈Pη0(PW)E
d
e(U, V )
and µ = δd
ewhere d
e= max
u,vd
e(u, v). Thus for all δ ≤ min(¯ δ
1, δ ¯
2, δ ¯
3) and n ≥ max(¯ n
1, n ¯
2, n ¯
3) we have
Pr
Q
Xn⋆∈ Q
δ( R + η
2, D + µ)
(56)
≥1 − Pr
Q
Xn⋆∈ / Q
−δ( R + η
2)
− Pr
Q
Xn⋆∈ / Q
◦δ( D + µ)
≥ 1 − ε
1− 2ε
2− ε
3. (57) This shows the existence of a strategy τ
⋆with codebook (w
n(m), x
n(m))
m∈{1,...,2nR}such that (57) is satisfied. We consider σ ∈ BR
e(τ
⋆) that achieves the maximum in (16) and we denote d
d= max
u,vd
d(u, v). The correspondance (δ, R , D ) 7→ Q
δ( R , D ) is continuous, from Berge’s Maximum Theorem we have
d
nd(σ, τ
⋆) = E
QXn⋆PV|W
h
d
d(U, V ) i
≤ sup
PU W∈ Qδ(R+η2,D+µ)
E
PU W PV|Wh
d
d(U, V ) i
+ (ε
1+ 2ε
2+ ε
3)d
d≤ sup
PU W∈ Q(R−η0,D)
E
PU W PV|Wh d
d(U, V ) i
+ (ε
1+ 2ε
2+ ε
3+ ε
4)d
d= max
PU W∈ Aη0
e (PW V)
E
PU W PV|Wh d
d(U, V ) i
+ (ε
1+ 2ε
2+ ε
3+ ε
4)d
d≤D
d⋆+ ε
0+ (ε
1+ 2ε
2+ ε
3+ ε
4)d
d.
We take ε
0, ε
1, ε
2, ε
3, ε
4, δ, η
2, η
0small and n ∈ N
⋆large
and the achievability result of Theorem 3 follows.
A
PPENDIXC P
ROOF OFL
EMMA1
Lemma 3 below ensures for all δ > 0, there exists a family of distributions (Q
kU W)
k∈K⊂ int ∆(U × W) with |K| < +∞
such that
∆(U × W) ⊂ [
k∈K
T
δ(Q
kU W), (58) min
k∈Kmin
(u,w)∈U ×W
Q
k(u, w) ≥ δ
4(|U × W| − 1) . (59) Thus for all δ > 0, there exists a family of distributions (Q
kU W˜)
˜k∈Ke⊂ Q
+δ( R + η) ∩ int ∆(U × W) with | K| e < +∞
such that (59) is satisfied and Q
+δ( R + η) ⊂ [
˜k∈Ke
T
δ(Q
kU W˜). (60) We choose δ < δ ¯ such that 3¯ δ log
4(|U ×W|−1)¯δ
< η.
Pr
∃m ∈ {1, . . . , 2
nR} s.t. Q
nm∈ Q
+δ( R + η)
(61)
≤Pr
∃m s.t. Q
nm∈ [
˜k∈Ke
T
δ(Q
˜kU W)
(62)
=Pr
∃ k ˜ ∈ K, e ∃m s.t. Q
nm∈ T
δ(Q
kU W˜)
(63)
≤ X
k∈˜ Ke
X
m∈{1,...,2nR}
X
(un ,wn)∈
Tδ(Q˜k U W)
P
U⊗n(u
n)P
W⊗n(w
n) (64)
≤| K| · e 2
n(R−I(U;W)+3δlog4(|U×W |−1)δ )
(65)
≤| K| · e 2
−n(η−3δlog4(|U×W |−1)δ )
. (66)
Equation (62) comes from (60). Equation (65) comes from (59) with min
u,wQ
k˜(u, w) ≥
4(|U ×W|−1)δ, and Proposition 2 and 3 below. Equation (66) comes from Q
˜kU W∈ Q
+δ( R + η), that induce R ≤ I(U ; W ) − η.
Since | K| e < +∞ and η − 3δ log
4(|U ×W|−1)δ
> 0, we choose n large such that | K| · e 2
−n(η−3δlog4(|U×W |−1)δ )
≤ ε.
This concludes the proof of Lemma 1.
Proposition 2 (see 1. pp. 27 in [15]) We consider P
U∈
∆(U), n ∈ N, δ > 0. For all u
n∈ T
δ(P
U) we have
2
−n(H(U)+δ1)≤ P
U⊗n(u
n) ≤ 2
−n(H(U)−δ1), (67) with δ
1= log
min1u∈suppPU
P(u)
· δ.
Proposition 3 (see 2. pp. 27 in [15]) We consider P
U W∈
∆(U × W), n ∈ N, δ > 0. Then T
δ(P
U W) ≤ 2
n(H(U,W)+δ2)with δ
2= log
min 1(u,w)∈suppPU W
P(u,w)
· δ.
Lemma 3 We consider a set U such that 2 ≤ |U| < +∞. For all δ > 0, there exists a family of distributions (Q
kU)
k∈K⊂ int ∆(U ) with |K| < +∞ such that
∆(U ) ⊂ [
k∈K
T
δ(Q
kU), min
k∈K
min
u∈U
Q
k(u) ≥ δ 4(|U| − 1) .
Proof. [Lemma 3] We consider a symbols u ˜ ∈ U and we define the distributions
P
U=
( 1 if U = ˜ u,
0 otherwise, Q
uU˜=
( 1 −
δ4if U = ˜ u,
δ
4(|U |−1)
otherwise.
Then,
||Q
uU˜− P
U||
1= X
u
|Q
u˜(u) − P(u)|
= δ
4 + δ
4(|U| − 1) (|U| − 1) = δ
2 < δ. (68) This shows that P
U∈ T
δ(Q
uU˜). The same construction applies to any other symbol u ˆ ∈ U , and this generates a collection of distributions (Q
uUˆ)
u∈Uˆ. We construct a family of distributions (Q
kU)
k∈K⊂ int ∆(U ) based on the lattice with steps
4(|U |−1)δthat connects the elements of (Q
uUˆ)
u∈Uˆ. Since
∆(U ) ⊂ [0, 1]
|U |−1, we have
|K| ≤ 4(|U| − 1) δ
|U |−1< +∞. (69)
R
EFERENCES[1] V. P. Crawford and J. Sobel, “Strategic information transmission,”
Econometrica, vol. 50, no. 6, pp. 1431–1451, 1982.
[2] E. Kamenica and M. Gentzkow, “Bayesian persuasion,”American Eco- nomic Review, vol. 101, pp. 2590–2615, 2011.
[3] M. Le Treust and T. Tomala, “Persuasion with limited communication capacity,”Journal of Economic Theory, vol. 184, p. 104940, 2019.
[4] S. Dughmi, R. Niazadeh, A. Psomas, and S. M. Weinberg, “Persuasion and incentives through the lens of duality,” inInternational Conference on Web and Internet Economics, pp. 142–155, Springer, 2019.
[5] S. Sarıta¸s, S. Yüksel, and S. Gezici, “Dynamic signaling games with quadratic criteria under Nash and Stackelberg equilibria,” Automatica, vol. 115, p. 108883, 2020.
[6] E. Akyol, C. Langbort, and T. Ba¸sar, “Information-theoretic approach to strategic communication as a hierarchical game,”Proceedings of the IEEE, vol. 105, no. 2, pp. 205–218, 2017.
[7] M. Le Treust and T. Tomala, “Information design for strategic coordi- nation of autonomous devices with non-aligned utilities,”IEEE Proc. of the 54th Allerton conference, Monticello, Illinois, pp. 233–242, 2016.
[8] M. Le Treust and T. Tomala, “Strategic coordination with state infor- mation at the decoder,” Proc. of 2018 International Zurich Seminar on Information and Communication, 2018.
[9] M. Le Treust and T. Tomala, “Strategic communication with side information at the decoder,” [on-line] available:
https://arxiv.org/abs/1911.04950, Nov. 2019.
[10] A. Lapidoth, “On the role of mismatch in rate distortion theory,”IEEE Transactions on Information Theory, vol. 43, pp. 38–47, Jan. 1997.
[11] J. Scarlett, A. G. i Fàbregas, A. Somekh-Baruch, and A. Martinez,
“Information-theoretic foundations of mismatched decoding,” Founda- tions and Trends in Communications and Information Theory, vol. 17, no. 2–3, pp. 149–401, 2020.
[12] A. S. Vora and A. A. Kulkarni, “Achievable rates for strategic communi- cation,” in2020 IEEE International Symposium on Information Theory (ISIT), pp. 1379–1384, 2020.
[13] A. S. Vora and A. A. Kulkarni, “Information extraction from a strategic sender: The zero error case,” [on-line] available:
https://arxiv.org/abs/2006.10641, 2020.
[14] M. O. Jackson and H. F. Sonnenschein, “Overcoming incentive con- straints by linking decisions,” Econometrica, vol. 75, pp. 241 – 257, January 2007.
[15] A. E. Gamal and Y.-H. Kim,Network Information Theory. Cambridge University Press, Dec. 2011.