Sample, computation vs storage tradeoffs for classification using tensor subspace models
This work addresses efficiency challenges in classification for applications requiring resource-constrained environments, presenting an incremental improvement over existing tensor subspace methods.
The paper tackles the tradeoffs between sample, computation, and storage complexity in supervised classification using tensor subspace models, showing that hierarchical Kronecker-structured subspaces improve these tradeoffs by preventing overfitting at higher latent dimensions.
In this paper, we exhibit the tradeoffs between the (training) sample, computation and storage complexity for the problem of supervised classification using signal subspace estimation. Our main tool is the use of tensor subspaces, i.e. subspaces with a Kronecker structure, for embedding the data into lower dimensions. Among the subspaces with a Kronecker structure, we show that using subspaces with a hierarchical structure for representing data leads to improved tradeoffs. One of the main reasons for the improvement is that embedding data into these hierarchical Kronecker structured subspaces prevents overfitting at higher latent dimensions.