The Relaxed Online Maximum Margin Algorithm

By Thomas G. Dietterich, S. Becker, Z. G. Eds. 1999

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

Fast Kernel Classifiers with Online and Active Learning (2005, unknown): “Other algorithms (Frieß et al., 1998; Gentile, 2001; Li and Long, 2002; Crammer and Singer, 2003) explicitely construct larger margins.”
Discovering physical concepts with neural networks (2018, methodology): “The concept of evolving the system in the latent representation has also been used in machine learning to extract the relevant features from video data [74].”
Accurate On-line Support Vector Regression (2003, unknown): “Approximate on-line training algorithms have previously been proposed for SVMs (Syed, Liu, & Sung, 1999; Csato & Opper, 2001; Gentile, 2001; Graepel, Herbrich, & Williamson, 2001; Herb-ster, 2001; Li & Long, 1999; Kivinen, Smola, & Williamson, 2002; Ralaivola & d’Alche-Buc, 2001).”
Breaking the curse of kernelization: budgeted stochastic gradient descent for large-scale SVM training (2012, unknown): “Popular recent members of this family of algorithms include ALMA (Gentile, 2001), ROMMA (Li and Long, 2002), MIRA (Crammer and Singer, 2003), PA (Crammer et al., 2006), ILK (Cheng et al., 2007), the SGD based algorithms (Kivinen et al., 2002; Zhang, 2004; Shalev-Shwartz et al., 2011), and the…”
Multistability and Perceptual Inference (2012, methodology): “ARTICLE Communicated by Nando de Freitas”
Entropy Minimization for Shadow Removal (2009, methodology): “3.2 we recover an approximate intrinsic RGB reflectivity, akin to that in (Tappen et al. 2003, 2005) but with a considerably less demanding algorithm: (Tappen et al. 2005) classified image gradients as illumination or reflectance edges depending on their direction and magnitude and, in cases of…”
Recent Advances in Adversarial Training for Adversarial Robustness (2021, methodology): “2, a series of methods [Alayrac et al., 2019; Carmon et al., 2019] leverage semi/unsupervised learning on unlabeled data to enhance the generalization of adversarial training.”
On handling negative transfer and imbalanced distributions in multiple source transfer learning (2014, unknown): “OL, which updates training model incrementally, has been extensively studied for years [19–23].”
Online Learning: A Comprehensive Survey (2018, unknown): “In literature, there are also some other first-order online learning algorithms, such as Approximate Large Margin Algorithms (ALMA) (Gentile, 2001) which is a large margin variant of the p-norm Perceptron algorithm, and the Relaxed Online Maximum Margin Algorithm (ROMMA) (Li and Long, 2002).”