Non-Uniform Stochastic Average Gradient Method for Training Conditional Random Fields

Conditional random fields (CRFs) [Lafferty et al., 2001] are a ubiquitous tool in natural language processing. They are used for part-of-speech tagging [McCallum et al., 2003], semantic role labeling [Cohn and Blunsom, 2005], topic modeling [Zhu and Xing, 2010], information extraction [Peng and McCallum, 2006], shallow parsing [Sha and Pereira, 2003], named-entity recognition [Settles, 2004], as well as a host of other applications in natural language processing and in other fields such as computer vision [Nowozin and Lampert, 2011]. Similar to generative Markov random field (MRF) models, CRFs allow us to model probabilistic dependencies between output variables. The key advantage of discriminative CRF models is the ability to use a very highdimensional feature set, without explicitly building a model for these features (as required by MRF models). Despite the widespread use of CRFs, a major disadvantage of these models is that they can be very slow to train and the time needed for numerical optimization in CRF models remains a bottleneck in many applications. Due to the high cost of evaluating the CRF objective function on even a single training example, it is now common to train CRFs using stochastic gradient methods [Vishwanathan et al., 2006]. These methods are advantageous over deterministic methods because on each iteration they only require computing the gradient of a single example (and not all example as in deterministic methods). Thus, if we have a data set with n training examples, the iterations of stochastic gradient methods are n times faster than deterministic methods. However, the number of stochastic gradient iterations required might be very high.