Equivariant and scale-free Tucker decomposition models
نویسنده
چکیده
Analyses of array-valued datasets often involve reduced-rank array approximations, typically obtained via least-squares or truncations of array decompositions. However, least-squares approximations tend to be noisy in high-dimensional settings, and may not be appropriate for arrays that include discrete or ordinal measurements. This article develops methodology to obtain low-rank model-based representations of continuous, discrete and ordinal data arrays. The model is based on a parameterization of the mean array as a multilinear product of a reduced-rank core array and a set of index-specific orthogonal eigenvector matrices. It is shown how orthogonally equivariant parameter estimates can be obtained from Bayesian procedures under invariant prior distributions. Additionally, priors on the core array are developed that act as regularizers, leading to improved inference over the standard least-squares estimator, and providing robustness to misspecification of the array rank. This model-based approach is extended to accommodate discrete or ordinal data arrays using a semiparametric transformation model. The resulting low-rank representation is scale-free, in the sense that it is invariant to monotonic transformations of the data array. In an example analysis of a multivariate discrete network dataset, this scale-free approach provides a more complete description of data patterns.
منابع مشابه
DinTucker: Scaling up Gaussian process models on multidimensional arrays with billions of elements
Infinite Tucker Decomposition (InfTucker) and random function prior models, as nonparametric Bayesian models on infinite exchangeable arrays, are more powerful models than widely-used multilinear factorization methods including Tucker and PARAFAC decomposition, (partly) due to their capability of modeling nonlinear relationships between array elements. Despite their great predictive performance...
متن کاملTensor Decompositions for Very Large Scale Problems
Modern applications such as neuroscience, text mining, and large-scale social networks generate massive amounts of data with multiple aspects and high dimensionality. Tensors (i.e., multi-way arrays) provide a natural representation for such massive data. Consequently, tensor decompositions and factorizations are emerging as novel and promising tools for exploratory analysis of multidimensional...
متن کاملBorel structurability on the 2-shift of a countable group
We show that for any infinite countable group G and for any free Borel action G y X there exists an equivariant class-bijective Borel map from X to the free part Free(2G) of the 2-shift G y 2G. This implies that any Borel structurability which holds for the equivalence relation generated by Gy Free(2G) must hold a fortiori for all equivalence relations coming from free Borel actions of G. A rel...
متن کاملDinTucker: Scaling Up Gaussian Process Models on Large Multidimensional Arrays
Tensor decomposition methods are effective tools for modelling multidimensional array data (i.e., tensors). Among them, nonparametric Bayesian models, such as Infinite Tucker Decomposition (InfTucker), are more powerful than multilinear factorization approaches, including Tucker and PARAFAC, and usually achieve better predictive performance. However, they are difficult to handle massive data du...
متن کامل2 00 9 Pieri resolutions for classical groups
We generalize the constructions of Eisenbud, Fløystad, and Weyman for equivariant minimal free resolutions over the general linear group, and we construct equivariant resolutions over the orthogonal and symplectic groups. We also conjecture and provide some partial results for the existence of an equivariant analogue of Boij–Söderberg decompositions for Betti tables, which were proven to exist ...
متن کامل