论文信息 - Optimal contraction order of multiple tensors

Optimal contraction order of multiple tensors

Computational problems involving big data require application of multiway analysis tools on modern parallel infrastructure. Tensor contraction is one the fundamental operations used in multiway analysis. Dynamic programming formulation presented in this work searches for the least memory using contraction order in multiple tensor contraction operation.

Ali Taylan Cemgil | Can Kavaklioglu

[1] David E. Bernholdt,et al. Synthesis of High-Performance Parallel Programs for a Class of ab Initio Quantum Chemistry Models , 2005, Proceedings of the IEEE.

[2] Ali Taylan Cemgil,et al. Generalised Coupled Tensor Factorisation , 2011, NIPS.

[3] P. Sadayappan,et al. Effective Utilization of Tensor Symmetry in Operation Optimization of Tensor Contraction Expressions , 2012 .

[4] Tamara G. Kolda,et al. Tensor Decompositions and Applications , 2009, SIAM Rev..

[5] Fumikazu Miwakeichi,et al. Concurrent EEG/fMRI analysis by multiway Partial Least Squares , 2004, NeuroImage.

[6] Chi-Chung Lam,et al. On Optimizing a Class of Multi-Dimensional Loops with Reductions for Parallel Execution , 1997, Parallel Process. Lett..

[7] David E. Bernholdt,et al. Automated Operation Minimization of Tensor Contraction Expressions in Electronic Structure Calculations , 2005, International Conference on Computational Science.

[8] P. Gács,et al. Algorithms , 1992 .

[9] Lieven De Lathauwer,et al. Fourth-Order Cumulant-Based Blind Identification of Underdetermined Mixtures , 2007, IEEE Transactions on Signal Processing.

[10] Demetri Terzopoulos,et al. Multilinear Analysis of Image Ensembles: TensorFaces , 2002, ECCV.