Learning CRFs with Hierarchical Features: An Application to Go
نویسندگان
چکیده
We investigate the task of learning grid-based CRFs with hierarchical features motivated by the task of territory prediction in Go. We first analyze various independent and gridbased CRF classification models and state-ofthe-art training/inference algorithms to determine which offers the best performance across a variety of metrics. Faced with the performance drawbacks of independent models and the computational drawbacks of intractable CRF models, we introduce the BMA-Tree algorithm that uses Bayesian model averaging of tree-structured predictors to exploit hierarchical feature structure. Our results demonstrate that BMA-Tree is superior to other independent classifiers and provides a computationally efficient alternative to intractable grid-based CRF models when training is too slow or approximate inference is inadequate for the task at hand.
منابع مشابه
Model-Guided Segmentation and Layout Labelling of Document Images Using a Hierarchical Conditional Random Field
We present a model-guided segmentation and document layout extraction scheme based on hierarchical Conditional Random Fields (CRFs, hereafter). Common methods to classify a pixel of a document image into classes text, background and image are often noisy, and error-prone, often requiring post-processing through heuristic methods. The input to the system is a pixel-wise classification based on t...
متن کاملSemi-Markov Conditional Random Field with High-Order Features
We extend first-order semi-Markov conditional random fields (semi-CRFs) to include higherorder semi-Markov features, and present efficient inference and learning algorithms, under the assumption that the higher-order semiMarkov features are sparse. We empirically demonstrate that high-order semi-CRFs outperform high-order CRFs and first-order semi-CRFs on three sequence labeling tasks with long...
متن کاملHigh-Dimensional Unsupervised Active Learning Method
In this work, a hierarchical ensemble of projected clustering algorithm for high-dimensional data is proposed. The basic concept of the algorithm is based on the active learning method (ALM) which is a fuzzy learning scheme, inspired by some behavioral features of human brain functionality. High-dimensional unsupervised active learning method (HUALM) is a clustering algorithm which blurs the da...
متن کاملFour Applications of Graphical Models in Machine Learning
Conditional random fields for multi-agent reinforcement learning Conditional random fields (CRFs, [1]) are graphical models for modeling the probability of labels given the observations. They have traditionally been trained with using a set of observation and label pairs. Underlying all CRFs is the assumption that, conditioned on the training data, the labels are independent and identically dis...
متن کاملA Simple and Efficient Model Pruning Method for Conditional Random Fields
Conditional random fields (CRFs) have been quite successful in various machine learning tasks. However, as larger and larger data become acceptable for the current computational machines, trained CRFs Models for a real application quickly inflate. Recently, researchers often have to use models with tens of millions features. This paper considers pruning an existing CRFs model for storage reduct...
متن کامل