Page 206 - Neural Network Modeling and Identification of Dynamical Systems

REFERENCES 197
[22] Pearlmutter BA. Learning state space trajectories in re- [39] Suykens JAK, Vandewalle J. Learning a simple recurrent
current neural networks. In: International 1989 Joint neural state space model to behave like Chua’s double
Conference on Neural Networks, vol. 2; 1989. p. 365–72. scroll. IEEE Trans Circuits Syst I, Fundam Theory Appl
[23] Sato MA. A real time learning algorithm for recurrent 1995;42(8):499–502.
analog neural networks. Biol Cybern 1990;62(3):237–41. [40] Bengio Y, Louradour J, Collobert R, Weston J. Curricu-
[24] Özyurt DB, Barton PI. Cheap second order directional lum learning. In: Proceedings of the 26th Annual Inter-
derivatives of stiff ODE embedded functionals. SIAM J national Conference on Machine Learning, ICML ’09.
Sci Comput 2005;26(5):1725–43. New York, NY, USA: ACM. ISBN 978-1-60558-516-1,
[25] Griewank A, Walther A. Evaluating derivatives: Princi- 2009. p. 41–8.
ples and techniques of algorithmic differentiation. 2nd [41] Fedorov VV. Theory of optimal experiments. New York:
ed. Philadelphia, PA, USA: Society for Industrial and Academic Press; 1972.
Applied Mathematics. ISBN 0898716594, 2008. [42] MacKay DJC. Information-based objective functions for
[26] CppAD, a package for differentiation of C++ algo- active data selection. Neural Comput 1992;4(4):590–604.
rithms. https://www.coin-or.org/CppAD/. [43] Cohn DA. Neural network exploration using optimal
[27] Walther A, Griewank A. Getting started with ADOL-C. experiment design. Neural Netw 1996;9(6):1071–83.
In: Naumann U, Schenk O, editors. Combinatorial sci- [44] Póczos B, Lörincz A. Identification of recurrent neural
entific computing. Chapman-Hall CRC computational networks by Bayesian interrogation techniques. J Mach
science; 2012. p. 181–202. Chap. 7.
Learn Res 2009;10:515–54.
[28] Allgower E, Georg K. Introduction to numerical con- [45] Shewry MC, Wynn HP. Maximum entropy sampling. J
tinuation methods. Philadelphia, PA, USA: Society for Appl Stat 1987;14(2):165–70.
Industrial and Applied Mathematics. ISBN 089871544X, [46] Wynn HP. Maximum entropy sampling and general
2003.
[29] Shalashilin VI, Kuznetsov EB. Parametric continuation equivalence theory. In: Di Bucchianico A, Läuter H,
WynnHP,editors.mODa7—Advancesinmodel-
and optimal parametrization in applied mathematics oriented design and analysis. Heidelberg: Physica-
and mechanics. Dordrecht, Boston, London: Kluwer Verlag HD; 2004. p. 211–8.
Academic Publishers; 2003.
[30] Chow SN, Mallet-Paret J, Yorke JA. Finding zeros of [47] Kozachenko L, Leonenko N. Sample estimate of
the entropy of a random vector. Probl Inf Transm
maps: Homotopy methods that are constructive with 1987;23:95–101.
probability one. Math Comput 1978;32:887–99.
[48] Kennedy J, Eberhart R. Particle swarm optimization. In:
[31] Watson LT. Theory of globally convergent probability-
Proceedings of ICNN’95 – IEEE International Confer-
one homotopies for nonlinear programming. SIAM J
ence on Neural Networks, vol. 4. ISBN 0-7803-2768-3,
Optim 2000;11(3):761–80.
1995. p. 1942–8.
[32] Chow J, Udpa L, Udpa SS. Homotopy continuation
methods for neural networks. In: IEEE International [49] van den Bergh F, Engelbrecht A. A new locally conver-
Symposium on Circuits and Systems, vol. 5; 1991. gent particle swarm optimiser. In: IEEE International
p. 2483–6. Conference on Systems, Man and Cybernetics, vol. 3;
[33] Lendl M, Unbehauen R, Luo FL. A homotopy 2002. p. 6.
method for training neural networks. Signal Process [50] Peer ES, van den Bergh F, Engelbrecht AP. Using neigh-
1998;64(3):359–70. bourhoods with the guaranteed convergence PSO. In:
[34] Gorse D, Shepherd AJ, Taylor JG. The new era in super- Proceedings of the SIS ’03 – IEEE Swarm Intelligence
vised learning. Neural Netw 1997;10(2):343–52. Symposium; 2003. p. 235–42.
[35] Coetzee FM. Homotopy approaches for the analysis and [51] Clerc M. Particle swarm optimization. Newport Beach,
solution of neural network and other nonlinear systems CA, USA: ISTE. ISBN 9781905209040, 2010.
of equations. Ph.D. thesis, Carnegie Mellon University; [52] Hansen N, Ostermeier A. Adapting arbitrary nor-
1995. mal mutation distributions in evolution strategies:
[36] Allgower EL, Georg K. Numerical path following. In: The covariance matrix adaptation. In: Proceedings of
Techniques of scientific computing (Part 2). Handbook the IEEE Conference on Evolutionary Computation.
of numerical analysis, vol. 5. Elsevier; 1997. p. 3–207. ISBN 0-7803-2902-3, 1996. p. 312–7.
[37] Elman JL. Learning and development in neural net- [53] Hansen N, Ostermeier A. Completely derandomized
works: the importance of starting small. Cognition self-adaptation in evolution strategies. Evol Comput
1993;48(1):71–99. 2001;9(2):159–95.
[38] Ludik J, Cloete I. Incremental increased complexity [54] Jastrebski G, Arnold D. Improving evolution strate-
training. In: Proc. ESANN 1994, 2nd European Sym. on gies through active covariance matrix adaptation. Evol
Artif. Neural Netw.; 1994. p. 161–5. Comput 2006:2814–21.