Results 1  10
of
310
Learning to detect unseen object classes by betweenclass attribute transfer
 In CVPR
, 2009
"... We study the problem of object classification when training and test classes are disjoint, i.e. no training examples of the target classes are available. This setup has hardly been studied in computer vision research, but it is the rule rather than the exception, because the world contains tens of t ..."
Abstract

Cited by 359 (5 self)
 Add to MetaCart
(Show Context)
We study the problem of object classification when training and test classes are disjoint, i.e. no training examples of the target classes are available. This setup has hardly been studied in computer vision research, but it is the rule rather than the exception, because the world contains tens of thousands of different object classes and for only a very few of them image, collections have been formed and annotated with suitable class labels. In this paper, we tackle the problem by introducing attributebased classification. It performs object detection based on a humanspecified highlevel description of the target objects instead of training images. The description consists of arbitrary semantic attributes, like shape, color or even geographic information. Because such properties transcend the specific learning task at hand, they can be prelearned, e.g. from image datasets unrelated to the current task. Afterwards, new classes can be detected based on their attribute representation, without the need for a new training phase. In order to evaluate our method and to facilitate research in this area, we have assembled a new largescale dataset, “Animals with Attributes”, of over 30,000 animal images that match the 50 classes in Osherson’s classic table of how strongly humans associate 85 semantic attributes with animal classes. Our experiments show that by using an attribute layer it is indeed possible to build a learning object detection system that does not require any training images of the target classes. 1.
What does classifying more than 10,000 image categories tell us?
"... Image classification is a critical task for both humans and computers. One of the challenges lies in the large scale of the semantic space. In particular, humans can recognize tens of thousands of object classes and scenes. No computer vision algorithm today has been tested at this scale. This pape ..."
Abstract

Cited by 116 (12 self)
 Add to MetaCart
(Show Context)
Image classification is a critical task for both humans and computers. One of the challenges lies in the large scale of the semantic space. In particular, humans can recognize tens of thousands of object classes and scenes. No computer vision algorithm today has been tested at this scale. This paper presents a study of large scale categorization including a series of challenging experiments on classification with more than 10, 000 image classes. We find that a) computational issues become crucial in algorithm design; b) conventional wisdom from a couple of hundred image categories on relative performance of different classifiers does not necessarily hold when the number of categories increases; c) there is a surprisingly strong relationship between the structure of WordNet (developed for studying language) and the difficulty of visual categorization; d) classification can be improved by exploiting the semantic hierarchy. Toward the future goal of developing automatic vision algorithms to recognize tens of thousands or even millions of image categories, we make a series of observations and arguments about dataset scale, category density, and image hierarchy.
MultiLabel Prediction via Compressed Sensing
, 902
"... We consider multilabel prediction problems with large output spaces under the assumption of output sparsity – that the target vectors have small support. We develop a general theory for a variant of the popular ECOC (error correcting output code) scheme, based on ideas from compressed sensing for e ..."
Abstract

Cited by 96 (3 self)
 Add to MetaCart
(Show Context)
We consider multilabel prediction problems with large output spaces under the assumption of output sparsity – that the target vectors have small support. We develop a general theory for a variant of the popular ECOC (error correcting output code) scheme, based on ideas from compressed sensing for exploiting this sparsity. The method can be regarded as a simple reduction from multilabel regression problems to binary regression problems. It is shown that the number of subproblems need only be logarithmic in the total number of label values, making this approach radically more efficient than others. We also state and prove performance guarantees for this method, and test it empirically. 1.
Label Ranking by Learning Pairwise Preferences
"... Preference learning is an emerging topic that appears in different guises in the recent literature. This work focuses on a particular learning scenario called label ranking, where the problem is to learn a mapping from instances to rankings over a finite number of labels. Our approach for learning s ..."
Abstract

Cited by 87 (20 self)
 Add to MetaCart
(Show Context)
Preference learning is an emerging topic that appears in different guises in the recent literature. This work focuses on a particular learning scenario called label ranking, where the problem is to learn a mapping from instances to rankings over a finite number of labels. Our approach for learning such a mapping, called ranking by pairwise comparison (RPC), first induces a binary preference relation from suitable training data using a natural extension of pairwise classification. A ranking is then derived from the preference relation thus obtained by means of a ranking procedure, whereby different ranking methods can be used for minimizing different loss functions. In particular, we show that a simple (weighted) voting strategy minimizes risk with respect to the wellknown Spearman rank correlation. We compare RPC to existing label ranking methods, which are based on scoring individual labels instead of comparing pairs of labels. Both empirically and theoretically, it is shown that RPC is superior in terms of computational efficiency, and at least competitive in terms of accuracy.
Label Embedding Trees for Large MultiClass Tasks
"... Multiclass classification becomes challenging at test time when the number of classes is very large and testing against every possible class can become computationally infeasible. This problem can be alleviated by imposing (or learning) a structure over the set of classes. We propose an algorithm f ..."
Abstract

Cited by 85 (2 self)
 Add to MetaCart
(Show Context)
Multiclass classification becomes challenging at test time when the number of classes is very large and testing against every possible class can become computationally infeasible. This problem can be alleviated by imposing (or learning) a structure over the set of classes. We propose an algorithm for learning a treestructure of classifiers which, by optimizing the overall tree loss, provides superior accuracy to existing tree labeling methods. We also propose a method that learns to embed labels in a low dimensional space that is faster than nonembedding approaches and has superior accuracy to existing embedding approaches. Finally we combine the two ideas resulting in the label embedding tree that outperforms alternative methods including OnevsRest while being orders of magnitude faster. 1
Statistical analysis of some multicategory large margin classification methods
 Journal of Machine Learning Research
, 2004
"... The purpose of this paper is to investigate statistical properties of risk minimization based multicategory classification methods. These methods can be considered as natural extensions of binary large margin classification. We establish conditions that guarantee the consistency of classifiers obtai ..."
Abstract

Cited by 72 (2 self)
 Add to MetaCart
The purpose of this paper is to investigate statistical properties of risk minimization based multicategory classification methods. These methods can be considered as natural extensions of binary large margin classification. We establish conditions that guarantee the consistency of classifiers obtained in the risk minimization framework with respect to the classification error. Examples are provided for four specific forms of the general formulation, which extend a number of known methods. Using these examples, we show that some risk minimization formulations can also be used to obtain conditional probability estimates for the underlying problem. Such conditional probability information can be useful for statistical inferencing tasks beyond classification. 1.
Approximations for Binary Gaussian Process Classification
"... We provide a comprehensive overview of many recent algorithms for approximate inference in Gaussian process models for probabilistic binary classification. The relationships between several approaches are elucidated theoretically, and the properties of the different algorithms are corroborated by ex ..."
Abstract

Cited by 59 (2 self)
 Add to MetaCart
We provide a comprehensive overview of many recent algorithms for approximate inference in Gaussian process models for probabilistic binary classification. The relationships between several approaches are elucidated theoretically, and the properties of the different algorithms are corroborated by experimental results. We examine both 1) the quality of the predictive distributions and 2) the suitability of the different marginal likelihood approximations for model selection (selecting hyperparameters) and compare to a gold standard based on MCMC. Interestingly, some methods produce good predictive distributions although their marginal likelihood approximations are poor. Strong conclusions are drawn about the methods: The Expectation Propagation algorithm is almost always the method of choice unless the computational budget is very tight. We also extend existing methods in various ways, and provide unifying code implementing all approaches. Keywords: Gaussian process priors, probabilistic classification, Laplaces’s approximation, expectation propagation, variational bounding, mean field methods, marginal likelihood evidence,
Extracting Shared Subspace for Multilabel Classification
"... Multilabel problems arise in various domains such as multitopic document categorization and protein function prediction. One natural way to deal with such problems is to construct a binary classifier for each label, resulting in a set of independent binary classification problems. Since the multipl ..."
Abstract

Cited by 54 (2 self)
 Add to MetaCart
(Show Context)
Multilabel problems arise in various domains such as multitopic document categorization and protein function prediction. One natural way to deal with such problems is to construct a binary classifier for each label, resulting in a set of independent binary classification problems. Since the multiple labels share the same input space, and the semantics conveyed by different labels are usually correlated, it is essential to exploit the correlation information contained in different labels. In this paper, we consider a general framework for extracting shared structures in multilabel classification. In this framework, a common subspace is assumed to be shared among multiple labels. We show that the optimal solution to the proposed formulation can be obtained by solving a generalized eigenvalue problem, though the problem is nonconvex. For highdimensional problems, direct computation of the solution is expensive, and we develop an efficient algorithm for this case. One appealing feature of the proposed framework is that it includes several wellknown algorithms as special cases, thus elucidating their intrinsic relationships. We have conducted extensive experiments on eleven multitopic web page categorization tasks, and results demonstrate the effectiveness of the proposed formulation in comparison with several representative algorithms.
Good Practice in LargeScale Learning for Image Classification
 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (TPAMI)
, 2013
"... We benchmark several SVM objective functions for largescale image classification. We consider onevsrest, multiclass, ranking, and weighted approximate ranking SVMs. A comparison of online and batch methods for optimizing the objectives shows that online methods perform as well as batch methods i ..."
Abstract

Cited by 51 (7 self)
 Add to MetaCart
(Show Context)
We benchmark several SVM objective functions for largescale image classification. We consider onevsrest, multiclass, ranking, and weighted approximate ranking SVMs. A comparison of online and batch methods for optimizing the objectives shows that online methods perform as well as batch methods in terms of classification accuracy, but with a significant gain in training speed. Using stochastic gradient descent, we can scale the training to millions of images and thousands of classes. Our experimental evaluation shows that rankingbased algorithms do not outperform the onevsrest strategy when a large number of training examples are used. Furthermore, the gap in accuracy between the different algorithms shrinks as the dimension of the features increases. We also show that learning through crossvalidation the optimal rebalancing of positive and negative examples can result in a significant improvement for the onevsrest strategy. Finally, early stopping can be used as an effective regularization strategy when training with online algorithms. Following these “good practices”, we were able to improve the stateoftheart on a large subset of 10K classes and 9M images of ImageNet from 16.7 % Top1 accuracy to 19.1%.
Solving Multiclass Support Vector Machines with LaRank
 In 24th International Conference on Machine Learning
, 2007
"... Optimization algorithms for large margin multiclass recognizers are often too costly to handle ambitious problems with structured outputs and exponential numbers of classes. Optimization algorithms that rely on the full gradient are not effective because, unlike the solution, the gradient is not spa ..."
Abstract

Cited by 45 (3 self)
 Add to MetaCart
(Show Context)
Optimization algorithms for large margin multiclass recognizers are often too costly to handle ambitious problems with structured outputs and exponential numbers of classes. Optimization algorithms that rely on the full gradient are not effective because, unlike the solution, the gradient is not sparse and is very large. The LaRank algorithm sidesteps this difficulty by relying on a randomized exploration inspired by the perceptron algorithm. We show that this approach is competitive with gradient based optimizers on simple multiclass problems. Furthermore, a single LaRank pass over the training examples delivers test error rates that are nearly as good as those of the final solution. 1.