Constraint Reasoning and Kernel Clustering for Pattern Decomposition with Scaling
Motivated by an important and challenging task encountered in material discovery, we consider the problem of finding K basis patterns of numbers that jointly compose N observed patterns while enforcing additional spatial and scaling constraints. We propose a Constraint Programming (CP) model which captures the exact problem structure yet fails to scale in the presence of noisy data about the patterns. We alleviate this issue by employing Machine Learning (ML) techniques, namely kernel methods and clustering, to decompose the problem into smaller ones based on a global data-driven view, and then stitch the partial solutions together using a global CP model. Combining the complementary strengths of CP and ML techniques yields a more accurate and scalable method than the few found in the literature for this complex problem.
KeywordsConstraint Programming Dynamic Time Warping Basis Pattern Pattern Decomposition Missing Element
Unable to display preview. Download preview PDF.
- 1.Powder Diffract. File, JCPDS Internat. Centre Diffract. Data, PA (2004)Google Scholar
- 6.Damoulas, T., Henry, S., Farnsworth, A., Lanzone, M., Gomes, C.: Bayesian Classification of Flight Calls with a novel Dynamic Time Warping Kernel. In: ICMLA 2010, pp. 424–429. IEEE, Los Alamitos (2010)Google Scholar
- 7.Gervet, C., Hentenryck, P.V.: Length-lex ordering for set csps. In: AAAI (2006)Google Scholar
- 8.Gomes, C.P.: Computational Sustainability: Computational methods for a sustainable environment, economy, and society. The Bridge, NAE 39(4) (2009)Google Scholar
- 13.Holm, J., de Lichtenberg, K., Thorup, M.: Poly-logarithmic deterministic fully-dynamic algorithms for connectivity, minimum spanning tree, 2-edge, and biconnectivity. In: STOC 1998, New York, NY, USA, pp. 79–89 (1998)Google Scholar
- 14.Long, C.J., Bunker, D., Karen, V.L., Li, X., Takeuchi, I.: Rapid identification of structural phases in combinatorial thin-film libraries using x-ray diffraction and non-negative matrix factorization. Rev. Sci. Instruments 80(103902) (2009)Google Scholar
- 15.Long, C.J., Hattrick-Simpers, J., Murakami, M., Srivastava, R.C., Takeuchi, I., Karen, V.L., Li, X.: Rapid structural mapping of ternary metallic alloy systems using the combinatorial approach and cluster analysis. Rev. Sci. Inst. 78 (2007)Google Scholar
- 16.Potyrailo, R.A., Maier, W.F.: Combinatorial and High-Throughput Discovery and Optimization of Catalysts and Materials. CRC Press, Boca Raton (2007)Google Scholar
- 17.Prosser, P., Unsworth, C.: A connectivity constraint using bridges. In: ECAI 2006, The Netherlands, pp. 707–708 (2006)Google Scholar
- 18.Sakoe, H., Chiba, S.: Dynamic programming algorithm optimization for spoken word recognition. Readings in Speech Recognition, 159 (1990)Google Scholar
- 19.Stockmeyer, L.J.: The set basis problem is np-complete. Technical Report Report No. RC-5431, IBM Watson Research Center, East Lansing, Michigan (1975)Google Scholar