![](bg6.png)
Table 2: TukeyHSD Test Results for Accuracy.
DE - sDE DE - sDispDE sDE - sDispDE
f
1
– – –
f
2
– X–o X–o
f
3
– – –
f
4
– – o–X
f
5
o–X o–X o–X
f
6
o–X o–X –
f
7
X–o – o–X
f
8
o–X o–X –
f
9
o–X o–X –
f
10
o–X o–X –
f
11
o–X o–X –
f
12
o–X o–X X–o
f
13
o–X o–X X–o
f
14
o–X o–X X–o
Based on TukeyHSD Test, if the difference between each pair of algorithms
is significant, the pairs are marked. X–o shows that the left algorithm is
significantly better than the right one; and o–X shows that the right algorithm
is significantly better than the one, on the left.
Algorithm 3: SDS Dispensation coupled with DE
(sDispDE).
// DIFFUSION PHASE
For ag = 1 to No_of_agents
If ( ag is not active )
ag.setHypo( randomHypo () )
Else
ag.setHypo(Gaussian (ag.getHypo (),aErrorV ))
End If
End For
7.1 Conclusions
This paper presents a brief overview about the poten-
tial of coupling of DE with SDS. Here, SDS is pri-
marily used as an efficient resource allocation and
dispensation mechanism responsible for facilitating
communication between the agents at the early stages
of the optimisation. Results reported in this paper
have demonstrated that initial explorations with the
coupled sDE algorithm outperform the performance
of (one variation of) classical DE architecture. We
believe similar techniques (e.g. (Omran et al., 2011))
can be applied to other swarm intelligence and evo-
lutionary algorithms. As reported in (al-Rifaie et al.,
2011a; al-Rifaie et al., 2011b) SDS has been also suc-
cessfully integrated (vs. coupled) into PSO and DE
in a different framework. In ongoing research, fur-
ther theoretical work seeks to develop the core ideas
presented in this paper on problems with significantly
more computationally expensive objective functions.
This reinforces the idea of the integration of SI
algorithms with EAs as a potential future approach in
Evolutionary Computation.
REFERENCES
al-Rifaie, M. M. and Bishop, M. (2010). The mining game:
a brief introduction to the stochastic diffusion search
metaheuristic. AISB Quarterly.
al-Rifaie, M. M., Bishop, M., and Blackwell, T. (2011a). An
investigation into the merger of stochastic diffusion
search and particle swarm optimisation. In GECCO
’11: Proceedings of the 2011 GECCO conference
companion on Genetic and evolutionary computation,
pages 37–44, New York, NY, USA. ACM.
al-Rifaie, M. M., Bishop, M., and Blackwell, T. (2011b).
Resource allocation and dispensation impact of
stochastic diffusion search on differential evolution al-
gorithm; in nature inspired cooperative strategies for
optimisation (NICSO 2011) proceedings. Studies in
Computational Intelligence. Springer.
Bishop, J. (1989). Stochastic searching networks. pages
329–331, London, UK. Proc. 1st IEE Conf. on Artifi-
cial Neural Networks.
Brest, J., Zamuda, A., Boskovic, B., Maucec, M., and
Zumer, V. (2009). Dynamic optimization using self-
adaptive differential evolution. In IEEE Congress
on Evolutionary Computation, 2009. CEC’09., pages
415–422. IEEE.
Nasuto, S. J. (1999). Resource Allocation Analysis of the
Stochastic Diffusion Search. PhD thesis, University
of Reading, Reading, UK.
Omran, M., Moukadem, I., al-Sharhan, S., and Kinawi,
M. (2011). Stochastic diffusion search for continu-
ous global optimization. International Conference on
Swarm Intelligence (ICSI 2011), Cergy, France.
Storn, R. and Price, K. (1995). Differential
evolution - a simple and efficient adaptive
scheme for global optimization over continu-
ous spaces. TR-95-012, [online]. Available:
http://www.icsi.berkeley.edu/ storn/litera.html.
Weber, M., Neri, F., and Tirronen, V. (2010). Parallel Ran-
dom Injection Differential Evolution. Applications of
Evolutionary Computation, pages 471–480.
Whitaker, R. and Hurley, S. (2002). An agent based ap-
proach to site selection for wireless networks. In
1st IEE Conf. on Artificial Neural Networks, Madrid
Spain. ACM Press Proc ACM Symposium on Applied
Computing.
Whitley, D., Rana, S., Dzubera, J., and Mathias, K. E.
(1996). Evaluating evolutionary algorithms. Artificial
Intelligence, 85(1-2):245–276.
Zaharie, D. (2003). Control of population diversity and
adaptation in differential evolution algorithms. In
Proc. of 9th International Conference on Soft Com-
puting, MENDEL, pages 41–46.
Zhang, J. and Sanderson, A. (2009). JADE: adaptive
differential evolution with optional external archive.
Evolutionary Computation, IEEE Transactions on,
13(5):945–958.
FEC 2011 - Special Session on Future of Evolutionary Computation
558