Title :
Regularization for matrix completion
Author :
Keshavan, Raghunandan H. ; Montanari, Andrea
Author_Institution :
Depts. of Electr. Eng., Stanford Univ., Stanford, CA, USA
Abstract :
We consider the problem of reconstructing a low rank matrix from noisy observations of a subset of its entries. This task has applications in statistical learning, computer vision, and signal processing. In these contexts, `noise´ generically refers to any contribution to the data that is not captured by the low-rank model. In most applications, the noise level is large compared to the underlying signal and it is important to avoid overfitting. In order to tackle this problem, we define a regularized cost function well suited for spectral reconstruction methods. Within a random noise model, and in the large system limit, we prove that the resulting accuracy undergoes a phase transition depending on the noise level and on the fraction of observed entries. The cost function can be minimized using OPTSPACE (a manifold gradient descent algorithm). Numerical simulations show that this approach is competitive with state-of-the-art alternatives.
Keywords :
computer vision; gradient methods; learning (artificial intelligence); matrix algebra; random noise; statistical analysis; OPTSPACE; computer vision; low rank matrix reconstruction; manifold gradient descent algorithm; matrix completion regularization; numerical simulations; phase transition; random noise model; regularized cost function; signal processing; spectral reconstruction methods; statistical learning; Application software; Computer vision; Context modeling; Cost function; Noise level; Phase noise; Reconstruction algorithms; Signal processing; Signal processing algorithms; Statistical learning;
Conference_Titel :
Information Theory Proceedings (ISIT), 2010 IEEE International Symposium on
Conference_Location :
Austin, TX
Print_ISBN :
978-1-4244-7890-3
Electronic_ISBN :
978-1-4244-7891-0
DOI :
10.1109/ISIT.2010.5513563