Information Technology Reference
In-Depth Information
Buyer Agent Seller Agent
Issue 1
min
1
=10
,max
1
=25
,w
1
=0
.
7
min
1
= 15,
max
1
= 30,
w
1
=0
.
5
Mixed Strategy (
γ
1
=0
.
3): Mixed Strategy (
γ
1
∈{
0
.
1
,
0
.
12
}
):
τ
1
,
time
: polynomial,
β
=5
τ
1
,
time
: polynomial,
β
=1
τ
1
,
beh
: absolute tft,
δ
=1
τ
1
,
beh
: absolute tft,
δ
=1
Issue 2
min
2
=20
,max
2
=40
,w
2
=0
.
3
min
2
=30
,max
2
=50
,w
2
=0
.
5
Mixed Strategy (
γ
1
=0
.
4): Mixed Strategy (
γ
1
=0
.
2):
τ
2
,
time
: polynomial,
β
=2
τ
2
,
time
: polynomial,
β
=0
.
3
τ
2
,
beh
: absolute tft,
δ
=1,
R
=0
τ
2
,
beh
:relativetft,
δ
=1
(a) Example 2: Negotiation settings
Iss
ue
1
Iss
ue
2
30
1.0
U
b
50
1.0
U
s
45
25
0.8
0.8
40
0.6
0.6
20
35
0.4
0.4
30
0.2
0.2
15
25
t
t
5
10
15
5
10
15
0.2
0.2
20
t
t
0
5
10
15
0
5
10
15
0.4
0.4
(b) Buyer's and seller's offer curves for is-
sue 1 and 2 (
γ
=0
.
12)
(c) Buyer's (left) and seller's (right) aggre-
gated offer utilities (
γ
=0
.
12)
Iss
ue
1
Iss
ue
2
1.0
U
b
30
50
1.0
U
s
45
25
0.8
0.8
40
0.6
0.6
20
35
0.4
0.4
30
0.2
0.2
15
25
t
t
5
10
15
5
10
15
0.2
0.2
20
t
t
0
5
10
15
0
5
10
15
0.4
0.4
(d) Buyer's and seller's offer curves for is-
sue 1 and 2 (
γ
=0
.
1)
(e) Buyer's (left) and seller's (right) aggre-
gated offer utilities (
γ
=0
.
1)
Fig. 1.
Example 2 settings and offer and utility curves using the traditional (straight)
or the negotiation thread-based mixing (dotted)
the sequence of opponents' offers is monotonic and all involved tactics are mono-
tonic as well. This is because the imitative tactic is
not independent
from the
other tactics in the mix since it uses the last offer of the current negotiation
thread which is different from the individually proposed one. In addition, if both
agents have imitative tactics in their mix a non-monotonic sequence of offers
is copied to some degree and may thus reproduce the non-monotonicity in the
sequence of opponent's offers and vice versa. If agents have opposing utility func-
tions, such that a non-monotonic utility sequence of one agent then also causes a
non-monotonic utility sequence of the partner's offers. This can result in a delay
of agreements, varying outcomes as compared to mixing methods with mono-
tonic offer sequences, and a high sensitivity in terms of the strategy parameters
making it dicult for an agent to apply such strategies in real world scenarios.
In such cases, the described dynamics of the system result in a partial loss of the
agent's control over its strategy since small changes of parameters may change
the offer curves to a large degree.
