Nonconvex Rectangular Matrix Completion via Gradient Descent without $\ell_{2,\infty}$ Regularization
This is an incremental improvement for the machine learning community, addressing computational efficiency in matrix completion without relying on ℓ2,∞ regularization.
The paper tackles the problem of nonconvex rectangular matrix completion by extending vanilla gradient descent analysis from positive semidefinite to rectangular cases, improving the required sampling rate from O(poly(κ)μ^3 r^3 log^3 n/n) to O(μ^2 r^2 κ^14 log n/n).
The analysis of nonconvex matrix completion has recently attracted much attention in the community of machine learning thanks to its computational convenience. Existing analysis on this problem, however, usually relies on $\ell_{2,\infty}$ projection or regularization that involves unknown model parameters, although they are observed to be unnecessary in numerical simulations, see, e.g., Zheng and Lafferty [2016]. In this paper, we extend the analysis of the vanilla gradient descent for positive semidefinite matrix completion proposed in Ma et al. [2017] to the rectangular case, and more significantly, improve the required sampling rate from $O(\operatorname{poly}(κ)μ^3 r^3 \log^3 n/n )$ to $O(μ^2 r^2 κ^{14} \log n/n )$. Our technical ideas and contributions are potentially useful in improving the leave-one-out analysis in other related problems.