Average-case linear matrix factorization and reconstruction of low width algebraic branching programs

A matrix X is called a linear matrix if its entries are affine forms, i.e., degree one polynomials in n variables. What is a minimal-sized representation of a given matrix F as a product of linear matrices? Finding such a minimal representation is closely related to finding an optimal way to compute...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Computational complexity Ročník 28; číslo 4; s. 749 - 828
Hlavní autori: Kayal, Neeraj, Nair, Vineet, Saha, Chandan
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Cham Springer International Publishing 01.12.2019
Springer Nature B.V
Predmet:
ISSN:1016-3328, 1420-8954
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:A matrix X is called a linear matrix if its entries are affine forms, i.e., degree one polynomials in n variables. What is a minimal-sized representation of a given matrix F as a product of linear matrices? Finding such a minimal representation is closely related to finding an optimal way to compute a given polynomial via an algebraic branching program. Here we devise an efficient algorithm for an average-case version of this problem. Specifically, given w , d , n ∈ N and blackbox access to the w 2 entries of a matrix product F = X 1 ⋯ X d , where each X i is a w × w linear matrix over a given finite field F q , we wish to recover a factorization F = Y 1 ⋯ Y d ′ , where every Y i is also a linear matrix over F q (or a small extension of F q ). We show that when the input F is sampled from a distribution defined by choosing random linear matrices X 1 , … , X d over F q independently and taking their product and n ≥ 4 w 2 and char ( F q ) = ( d n ) Ω ( 1 ) , then an equivalent factorization F = Y 1 ⋯ Y d can be recovered in (randomized) time ( d n log q ) O ( 1 ) . In fact, we give a (worst-case) polynomial time randomized algorithm to factor any non-degenerate or pure matrix product (a notion we define in the paper) into linear matrices; a matrix product F = X 1 ⋯ X d is pure with high probability when the X i 's are chosen independently at random. We also show that in this situation, if we are instead given a single entry of F rather than its w 2 correlated entries, then the recovery can be done in (randomized) time ( d w 3 n log q ) O ( 1 ) .
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1016-3328
1420-8954
DOI:10.1007/s00037-019-00189-0