Skip to yearly menu bar Skip to main content


Poster

Optimal Analysis of Subset-Selection Based L_p Low-Rank Approximation

Chen Dan · Hong Wang · Hongyang Zhang · Yuchen Zhou · Pradeep Ravikumar

East Exhibition Hall B, C #226

Keywords: [ Learning Theory ] [ Theory ]


Abstract: We show that for the problem of p rank-k approximation of any given matrix over Rn×m and Cn×m, the algorithm of column subset selection enjoys approximation ratio (k+1)1/p for 1p2 and (k+1)11/p for p2. This improves upon the previous O(k+1) bound (Chierichetti et al.,2017) for p1. We complement our analysis with lower bounds; these bounds match our upper bounds up to constant 1 when p2. At the core of our techniques is an application of Riesz-Thorin interpolation theorem from harmonic analysis, which might be of independent interest to other algorithmic designs and analysis more broadly. Our analysis results in improvements on approximation guarantees of several other algorithms with various time complexity. For example, to make the algorithm of column subset selection computationally efficient, we analyze a polynomial time bi-criteria algorithm which selects O(klogm) number of columns. We show that this algorithm has an approximation ratio of O((k+1)1/p) for 1p2 and O((k+1)11/p) for p2. This improves over the bound in (Chierichetti et al.,2017) with an O(k+1) approximation ratio. Our bi-criteria algorithm also implies an exact-rank method in polynomial time with a slightly larger approximation ratio.

Live content is unavailable. Log in and register to view live content