We propose a general algorithmic framework for constrained matrix and tensor factorization, which is widely used in unsupervised learning. The new framework is a hybrid between alternating optimization (AO) and the alternating direction method of multipliers (ADMM): each matrix factor is updated in turn, using ADMM. This combination can naturally accommodate a great variety of constraints on the factor matrices, hence the term 'universal'. Computation caching and warm start strategies are used to ensure that each update is evaluated efficiently, while the outer AO framework guarantees that the algorithm converges monotonically. Simulations on synthetic data show significantly improved performance relative to state-of-the-art algorithms.
|Original language||English (US)|
|Title of host publication||2015 23rd European Signal Processing Conference, EUSIPCO 2015|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|Number of pages||5|
|State||Published - Dec 22 2015|
|Event||23rd European Signal Processing Conference, EUSIPCO 2015 - Nice, France|
Duration: Aug 31 2015 → Sep 4 2015
|Name||2015 23rd European Signal Processing Conference, EUSIPCO 2015|
|Other||23rd European Signal Processing Conference, EUSIPCO 2015|
|Period||8/31/15 → 9/4/15|
Bibliographical noteFunding Information:
Supported in part by NSF IIS-1247632, IIS-1447788, and a UM Informatics Institute fellowship
© 2015 EURASIP.