Pith. sign in

Efficient Tensor Decomposition

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

This chapter studies the problem of decomposing a tensor into a sum of constituent rank one tensors. While tensor decompositions are very useful in designing learning algorithms and data analysis, they are NP-hard in the worst-case. We will see how to design efficient algorithms with provable guarantees under mild assumptions, and using beyond worst-case frameworks like smoothed analysis.

fields

cs.DS 1

years

2025 1

verdicts

ACCEPT 1

representative citing papers

citing papers explorer

Showing 1 of 1 citing paper.

  • Improving the Threshold for Finding Rank-1 Matrices in a Subspace cs.DS · 2025-04-24 · accept · none · ref 22 · internal anchor

    For generic planted rank-one matrices in an R-dimensional subspace of m by n matrices, the JLV algorithm is proven to recover them when R is about half of mn, and proven to fail above about 0.71 times mn.