Abstract | ||
---|---|---|
Various tensor decompositions use different arrangements of factors to explain multi-way data. Components from different decompositions can vary in the number of parameters. Allowing a model to contain components from different decompositions results in a combinatoric number of possible models. Model selection balances approximation error and the number of parameters, but due to the number of possible models, post-hoc model selection is infeasible. Instead, we incrementally build a model. This approach is analogous to sparse coding with a union of dictionaries. The proposed greedy approach can estimate a model consisting of a combination of tensor decompositions. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1109/ICASSP.2013.6638839 | 2013 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) |
Keywords | Field | DocType |
tensor decompositions, greedy algorithm, model | Mathematical optimization,Tensor,Computer science,Neural coding,Approximation theory,Model selection,Image coding,Greedy algorithm,Greedy randomized adaptive search procedure,Approximation error | Conference |
ISSN | Citations | PageRank |
1520-6149 | 1 | 0.37 |
References | Authors | |
10 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Austin J. Brockmeier | 1 | 20 | 5.24 |
Jose C. Principe | 2 | 2295 | 282.29 |
Anh Huy Phan | 3 | 828 | 51.60 |
Andrzej Cichocki | 4 | 5228 | 508.42 |