Cross-Entropy Optimization for Independent Process Analysis
Conference paper
- 9 Citations
- 2k Downloads
Abstract
We treat the problem of searching for hidden multi-dimensional independent auto-regressive processes. First, we transform the problem to Independent Subspace Analysis (ISA). Our main contribution concerns ISA. We show that under certain conditions, ISA is equivalent to a combinatorial optimization problem. For the solution of this optimization we apply the cross-entropy method. Numerical simulations indicate that the cross-entropy method can provide considerable improvements over other state-of-the-art methods.
Keywords
Independent Component Analysis Travelling Salesman Problem Independent Component Analysis Blind Source Separation Permutation Matrix
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Preview
Unable to display preview. Download preview PDF.
References
- 1.Choi, S., Cichocki, A., Park, H.M., Lee, S.Y.: Blind Source Separation and Independent Component Analysis. Neural Inf. Proc. Lett. and Reviews (2005)Google Scholar
- 2.Cardoso, J.: Multidimensional Independent Component Analysis. In: ICASSP 1998, Seattle, WA (1998)Google Scholar
- 3.Akaho, S., Kiuchi, Y., Umeyama, S.: MICA: Multimodal Independent Component Analysis. In: IJCNN, pp. 927–932 (1999)Google Scholar
- 4.Póczos, B., Takács, B., Lőrincz, A.: Independent Subspace Analysis on Innovations. In: Gama, J., Camacho, R., Brazdil, P.B., Jorge, A.M., Torgo, L. (eds.) ECML 2005. LNCS (LNAI), vol. 3720, pp. 698–706. Springer, Heidelberg (2005)CrossRefGoogle Scholar
- 5.Hyvärinen, A.: Independent Component Analysis for Time-dependent Stochastic Processes. In: ICANN 1998, pp. 541–546 (1998)Google Scholar
- 6.Vollgraf, R., Obermayer, K.: Multi-Dimensional ICA to Separate Correlated Sources. In: NIPS, vol. 14, pp. 993–1000 (2001)Google Scholar
- 7.Bach, F.R., Jordan, M.I.: Finding Clusters in Independent Component Analysis. In: ICA 2003, pp. 891–896 (2003)Google Scholar
- 8.Póczos, B., Lőrincz, A.: Independent Subspace Analysis Using k-Nearest Neighborhood Distances. In: Duch, W., Kacprzyk, J., Oja, E., Zadrożny, S. (eds.) ICANN 2005. LNCS, vol. 3697, pp. 163–168. Springer, Heidelberg (2005)Google Scholar
- 9.Póczos, B., Lőrincz, A.: Independent Subspace Analysis Using Geodesic Spanning Trees. In: ICML, pp. 673–680 (2005)Google Scholar
- 10.Theis, F.J.: Blind Signal Separation into Groups of Dependent Signals Using Joint Block Diagonalization. In: Proc. ISCAS 2005, Kobe, Japan, pp. 5878–5881 (2005)Google Scholar
- 11.Van Hulle, M.M.: Edgeworth Approximation of Multivariate Differential Entropy. Neural Comput. 17, 1903–1910 (2005)zbMATHCrossRefGoogle Scholar
- 12.Cheung, Y., Xu, L.: Dual Multivariate Auto-Regressive Modeling in State Space for Temporal Signal Separation. IEEE Tr. on Syst. Man Cyb. B 33, 386–398 (2003)Google Scholar
- 13.Theis, F.J.: Uniqueness of Complex andMultidimensional Independent Component Analysis. Signal Proc. 84, 951–956 (2004)zbMATHCrossRefGoogle Scholar
- 14.Szabó, Z., Póczos, B., Lőrincz, A.: Separation Theorem for Independent Subspace Analysis. Technical report, Eötvös Loránd University, Budapest (2005), http://people.inf.elte.hu/lorincz/Files/TR-ELU-NIPG-31-10-2005.pdf
- 15.Yukich, J.E.: Probability Theory of Classical Euclidean Optimization Problems. Lecture Notes in Math., vol. 1675. Springer, Berlin (1998)zbMATHGoogle Scholar
- 16.Costa, J.A., Hero, A.O.: Manifold Learning Using k-Nearest Neighbor Graphs. In: ICASSP, Montreal, Canada (2004)Google Scholar
- 17.Rubinstein, R.Y., Kroese, D.P.: The Cross-Entropy Method. Springer, Heidelberg (2004)zbMATHGoogle Scholar
- 18.Ekman, P.: Emotion in the Human Face. Cambridge Univ. Press, New York (1982)Google Scholar
- 19.Póczos, B., Lőrincz, A.: Non-combinatorial Estimation of Independent Autoregressive Sources (2005) (submitted)Google Scholar
Copyright information
© Springer-Verlag Berlin Heidelberg 2006