Reducing Multiclass to Binary: A Unifying Approach for Margin Classifiers

By Erin L. Allwein, Rob Schapire, Yoram Singer. 2000

1,693 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Multi-class texture analysis in colorectal cancer histology (2016, methodology): “We employed support vector machines (SVM) with one-versus-one class decisions in an error-correcting output code multiclass model (ECOC) 30,31 .”
Kernel methods in machine learning (2007, neutral): “[2], Collins [30] and Crammer and Singer [35]) can be encoded via Y = [N ], where N is the number of classes and ∆(y, y′) = 1 − δy,y′ .”
Sharing Visual Features for Multiclass and Multiview Object Detection (2007, methodology): “[2] show that the popular one-against-all approach is often suboptimal, but that the best code matrix is problem dependent.”
Multi-Label Prediction via Compressed Sensing (2009, background): “Th e loss based decoding approach [12] suggests decoding so as to minimize loss.”
On Oracle-Efficient PAC RL with Rich Observations (2018, methodology): “The oracle-based approach has produced practically effective algorithms for active learning [18], contextual bandits [19], structured prediction [20, 21], and multi-class classification [22], and here, we consider oracle-based algorithms for challenging RL settings.”
Beyond One-hot Encoding: lower dimensional target embedding (2018, methodology): “Indeed there are several approaches for generating ECOCs: Exhaustive codes [12], BCH codes [8], random codes [3], and circular ECOC [17] are few examples of methods that generate codes independently from the inherent structure of the data.”
Where am I in the dark: Exploring active transfer learning on the use of indoor localization based on thermal imaging (2016, background): “[46] proposed a framework to reduce k-class classification to l binary problem based on margin-based learning.”
Support Vector Learning for Semantic Argument Classification (2005, methodology): “There are two common approaches for extending SVMs to multi-class classification problems (Allwein et al., 2000).”
Boosting for Tumor Classification with Gene Expression Data (2003, background): “More sophisticated and computationally more expensive approaches for reducing multiclass to binary problems also exist, see Hastie and Tibshirani (1998) or Allwein et al. (2000) for a thorough discussion. The one-against-all approach allows for different preselected features, different chosen variab”
Sharing features: efficient boosting procedures for multiclass object detection (2004, background): “[1] show that the popular one-against-all approach is often suboptimal, but that the best code matrix to use is problem dependent.”
On the Learnability and Design of Output Codes for Multiclass Problems (2002, methodology): “Motivated by [24, 2] we seek a matrix M with a small norm which satisfies Eq.”
Automatic Tagging of Arabic Text: From Raw Text to Base Phrase Chunks (2004, unknown): “Thereby, we address them using the same SVM experimental setup which comprises a standard SVM as a multi-class classiﬁer (Allwein et al., 2000).”
Adaptive computation and machine learning (1998, background): “(2000) and Allwein et al. (2000). There exits a vast literature on support vector regression estimation; for an excellent overview see Smola and Schölkopf (2001), Smola (1996), Smola (1998) and Smola and Schölkopf (1998).”
Applications of support vector machines to speech recognition (2004, unknown): “On several standard classification tasks, it has been proven that one-versus-one classifiers are marginally more accurate than one-versus-all classifiers [16], [17].”
Designing Category-Level Attributes for Discriminative Visual Recognition (2013, result): “The above framework is motivated by two previous studies: learning attributes based on category-attribute matrix [13], and Error Correcting Output Code (ECOC) [1,5].”
Recent Advances of Large-Scale Linear Classification (2012, methodology): “Studies that broadly discussed various approaches of decomposition include, for example, [52] and [53].”