Results 1  10
of
234
Large margin methods for structured and interdependent output variables
 JOURNAL OF MACHINE LEARNING RESEARCH
, 2005
"... Learning general functional dependencies between arbitrary input and output spaces is one of the key challenges in computational intelligence. While recent progress in machine learning has mainly focused on designing flexible and powerful input representations, this paper addresses the complementary ..."
Abstract

Cited by 612 (12 self)
 Add to MetaCart
Learning general functional dependencies between arbitrary input and output spaces is one of the key challenges in computational intelligence. While recent progress in machine learning has mainly focused on designing flexible and powerful input representations, this paper addresses the complementary issue of designing classification algorithms that can deal with more complex outputs, such as trees, sequences, or sets. More generally, we consider problems involving multiple dependent output variables, structured output spaces, and classification problems with class attributes. In order to accomplish this, we propose to appropriately generalize the wellknown notion of a separation margin and derive a corresponding maximummargin formulation. While this leads to a quadratic program with a potentially prohibitive, i.e. exponential, number of constraints, we present a cutting plane algorithm that solves the optimization problem in polynomial time for a large class of problems. The proposed method has important applications in areas such as computational biology, natural language processing, information retrieval/extraction, and optical character recognition. Experiments from various domains involving different types of output spaces emphasize the breadth and generality of our approach.
Maxmargin Markov networks
, 2003
"... In typical classification tasks, we seek a function which assigns a label to a single object. Kernelbased approaches, such as support vector machines (SVMs), which maximize the margin of confidence of the classifier, are the method of choice for many such tasks. Their popularity stems both from the ..."
Abstract

Cited by 594 (15 self)
 Add to MetaCart
In typical classification tasks, we seek a function which assigns a label to a single object. Kernelbased approaches, such as support vector machines (SVMs), which maximize the margin of confidence of the classifier, are the method of choice for many such tasks. Their popularity stems both from the ability to use highdimensional feature spaces, and from their strong theoretical guarantees. However, many realworld tasks involve sequential, spatial, or structured data, where multiple labels must be assigned. Existing kernelbased methods ignore structure in the problem, assigning labels independently to each object, losing much useful information. Conversely, probabilistic graphical models, such as Markov networks, can represent correlations between labels, by exploiting problem structure, but cannot handle highdimensional feature spaces, and lack strong theoretical generalization guarantees. In this paper, we present a new framework that combines the advantages of both approaches: Maximum margin Markov (M 3) networks incorporate both kernels, which efficiently deal with highdimensional features, and the ability to capture correlations in structured data. We present an efficient algorithm for learning M 3 networks based on a compact quadratic program formulation. We provide a new theoretical bound for generalization in structured domains. Experiments on the task of handwritten character recognition and collective hypertext classification demonstrate very significant gains over previous approaches. 1
Support vector machine learning for interdependent and structured output spaces
 In ICML
, 2004
"... Learning general functional dependencies is one of the main goals in machine learning. Recent progress in kernelbased methods has focused on designing flexible and powerful input representations. This paper addresses the complementary issue of problems involving complex outputs suchas multiple depe ..."
Abstract

Cited by 444 (20 self)
 Add to MetaCart
(Show Context)
Learning general functional dependencies is one of the main goals in machine learning. Recent progress in kernelbased methods has focused on designing flexible and powerful input representations. This paper addresses the complementary issue of problems involving complex outputs suchas multiple dependent output variables and structured output spaces. We propose to generalize multiclass Support Vector Machine learning in a formulation that involves features extracted jointly from inputs and outputs. The resulting optimization problem is solved efficiently by a cutting plane algorithm that exploits the sparseness and structural decomposition of the problem. We demonstrate the versatility and effectiveness of our method on problems ranging from supervised grammar learning and namedentity recognition, to taxonomic text classification and sequence alignment. 1.
Online passiveaggressive algorithms
 JMLR
, 2006
"... We present a unified view for online classification, regression, and uniclass problems. This view leads to a single algorithmic framework for the three problems. We prove worst case loss bounds for various algorithms for both the realizable case and the nonrealizable case. The end result is new alg ..."
Abstract

Cited by 420 (24 self)
 Add to MetaCart
(Show Context)
We present a unified view for online classification, regression, and uniclass problems. This view leads to a single algorithmic framework for the three problems. We prove worst case loss bounds for various algorithms for both the realizable case and the nonrealizable case. The end result is new algorithms and accompanying loss bounds for hingeloss regression and uniclass. We also get refined loss bounds for previously studied classification algorithms.
Learning structured prediction models: a large margin approach
, 2004
"... We consider large margin estimation in a broad range of prediction models where inference involves solving combinatorial optimization problems, for example, weighted graphcuts or matchings. Our goal is to learn parameters such that inference using the model reproduces correct answers on the training ..."
Abstract

Cited by 225 (8 self)
 Add to MetaCart
(Show Context)
We consider large margin estimation in a broad range of prediction models where inference involves solving combinatorial optimization problems, for example, weighted graphcuts or matchings. Our goal is to learn parameters such that inference using the model reproduces correct answers on the training data. Our method relies on the expressive power of convex optimization problems to compactly capture inference or solution optimality in structured prediction models. Directly embedding this structure within the learning formulation produces concise convex problems for efficient estimation of very complex and diverse models. We describe experimental results on a matching task, disulfide connectivity prediction, showing significant improvements over stateoftheart methods. 1.
Training structural SVMs when exact inference is intractable
 IN: PROC. INTL. CONF. ON MACHINE LEARNING (ICML
, 2008
"... While discriminative training (e.g., CRF, structural SVM) holds much promise for machine translation, image segmentation, and clustering, the complex inference these applications require make exact training intractable. This leads to a need for approximate training methods. Unfortunately, knowledge ..."
Abstract

Cited by 139 (7 self)
 Add to MetaCart
(Show Context)
While discriminative training (e.g., CRF, structural SVM) holds much promise for machine translation, image segmentation, and clustering, the complex inference these applications require make exact training intractable. This leads to a need for approximate training methods. Unfortunately, knowledge about how to perform efficient and effective approximate training is limited. Focusing on structural SVMs, we provide and explore algorithms for two different classes of approximate training algorithms, which we call undergenerating (e.g., greedy) and overgenerating (e.g., relaxations) algorithms. We provide a theoretical and empirical analysis of both types of approximate trained structural SVMs, focusing on fully connected pairwise Markov random fields. We find that models trained with overgenerating methods have theoretic advantages over undergenerating methods, are empirically robust relative to their undergenerating brethren, and relaxed trained models favor nonfractional predictions from relaxed predictors.
Kernel Conditional Random Fields: Representation and Clique Selection
 IN ICML
, 2004
"... Kernel conditional random fields (KCRFs) are introduced as a framework for discriminative modeling of graphstructured data. A representer theorem for conditional graphical models is given which shows how kernel conditional random fields arise from risk minimization procedures defined using Me ..."
Abstract

Cited by 95 (5 self)
 Add to MetaCart
Kernel conditional random fields (KCRFs) are introduced as a framework for discriminative modeling of graphstructured data. A representer theorem for conditional graphical models is given which shows how kernel conditional random fields arise from risk minimization procedures defined using Mercer kernels on labeled graphs. A procedure for greedily selecting cliques in the dual representation is then proposed, which allows sparse representations. By incorporating kernels and implicit feature spaces into conditional graphical models, the framework enables semisupervised learning algorithms for structured data through the use of graph kernels.
A Review of Kernel Methods in Machine Learning
, 2006
"... We review recent methods for learning with positive definite kernels. All these methods formulate learning and estimation problems as linear tasks in a reproducing kernel Hilbert space (RKHS) associated with a kernel. We cover a wide range of methods, ranging from simple classifiers to sophisticate ..."
Abstract

Cited by 95 (4 self)
 Add to MetaCart
We review recent methods for learning with positive definite kernels. All these methods formulate learning and estimation problems as linear tasks in a reproducing kernel Hilbert space (RKHS) associated with a kernel. We cover a wide range of methods, ranging from simple classifiers to sophisticated methods for estimation with structured data.
Exploiting dictionaries in named entity extraction: Combining semimarkov extraction processes and data integration method
 In Proceedings of the ACM SIGKDD Conference
, 2004
"... We consider the problem of improving named entity recognition (NER) systems by using external dictionaries—more specifically, the problem of extending stateoftheart NER systems by incorporating information about the similarity of extracted entities to entities in an external dictionary. This is d ..."
Abstract

Cited by 93 (6 self)
 Add to MetaCart
(Show Context)
We consider the problem of improving named entity recognition (NER) systems by using external dictionaries—more specifically, the problem of extending stateoftheart NER systems by incorporating information about the similarity of extracted entities to entities in an external dictionary. This is difficult because most highperformance named entity recognition systems operate by sequentially classifying words as to whether or not they participate in an entity name; however, the most useful similarity measures score entire candidate names. To correct this mismatch we formalize a semiMarkov extraction process which relaxes the usual Markov assumptions. This process is based on sequentially classifying segments of several adjacent words, rather than single words. In addition to allowing a natural way of coupling NER and highperformance record linkage methods, this formalism also allows the direct use of other useful entitylevel features, and provides a more natural formulation of the NER problem than sequential word classification. Experiments in multiple domains show that the new model can substantially improve extraction performance, relative to previously published methods for using external dictionaries in NER.
Latent Hierarchical Structural Learning for Object Detection
, 2010
"... We present a latent hierarchical structural learning method for object detection. An object is represented by a mixture of hierarchical tree models where the nodes represent object parts. The nodes can move spatially to allow both local and global shape deformations. The models can be trained discri ..."
Abstract

Cited by 84 (7 self)
 Add to MetaCart
We present a latent hierarchical structural learning method for object detection. An object is represented by a mixture of hierarchical tree models where the nodes represent object parts. The nodes can move spatially to allow both local and global shape deformations. The models can be trained discriminatively using latent structural SVM learning, where the latent variables are the node positions and the mixture component. But current learning methods are slow, due to the large number of parameters and latent variables, and have been restricted to hierarchies with two layers. In this paper we describe an incremental concaveconvex procedure (iCCCP) which allows us to learn both two and three layer models efficiently. We show that iCCCP leads to a simple training algorithm which avoids complex multistage layerwise training, careful part selection, and achieves good performance without requiring elaborate initialization. We perform object detection using our learnt models and obtain performance comparable with stateoftheart methods when evaluated on challenging public PASCAL datasets. We demonstrate the advantages of three layer hierarchies – outperforming Felzenszwalb et al.’s two layer models on all 20 classes.