Y. Li and P. Long. The relaxed online maximum margin algorithm. Machine Learning, 46:361--387, 2002.  Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Online Bayes Point Machines - Harrington, Herbrich, Kivinen.. (Correct)
....some immunity to attribute noise and concept drift. There have been a few recent attempts to develop further online algorithms that achieve an approximation to the maximum margin. Kivinen et al. 3] studied the marginalised perceptron (and issues arising when it is kernelised) Li and Long [2] studied an algorithm they called ROMMA where if there is a mistake at the tth trial then w t 1 is the smallest element of the constrained set of w : w t w kw t k fw : y t (w x t 1)g, else w t 1 = w t . It has a similar mistake bound to the perceptron and is computationally ....
Y. Li. & P. Long, (2002) The relaxed online maximum margin algorithm. Machine Learning, 46(1-3):361-387.
Online Learning with Kernels - Kivinen, Smola, Williamson (2001) (5 citations) (Correct)
....will ensure a limited number of updates per iteration. However they can be computationally expensive since they require one matrix multiplication at each step. The size of the matrix is given by the number of kernel functions required at each step. Recently several algorithms have been proposed [5, 8, 6, 12] performing perceptron like updates for classification at each step. Some algorithms work only in the noise free case, others not for moving targets, and yet again others assume an upper bound on the complexity of the estimators. In the present paper we present a simple method which will allows ....
Y. Li and P.M. Long. The relaxed online maximum margin algorithm. In S. A. Solla, T. K. Leen, and K.-R. Muller, editors, Advances in Neural Information Processing Systems 12, pages 498--504. MIT Press, 1999.
A Second-Order Perceptron Algorithm - Cesa-Bianchi, Conconi, Gentile (2002) (Correct)
....tune this parameter. Our second order Perceptron algorithm might be seen as a new on line classi cation technology. As such, this technology could be combined with previous techniques, such as the shifting target technique [16, 2] and the approximate on line large margin technique (e.g. [20, 14]) We have run simple experiments on synthetic data to give some evidence of the theoretical properties of our algorithm. Currently, we are running extensive experiments on real world data, such as textual data. These data are known to have a spectral structure that is exploitable in ....
Li, Y., & Long, P. (2002). The relaxed online maximum margin algorithm. Machine Learning Journal, 46(1/3), 361-387.
Ultraconservative Online Algorithms for Multiclass Problems - Crammer, Singer (2001) (6 citations) (Correct)
....with the algorithmic approaches described in this paper, they were all designed for batch problems and were not analyzed in the mistake bound model. Another approach to the problem of designing an update rule which results in a linear classi er of a small norm was suggested by Li and Long [13]. The algorithm Li and Long proposed, called ROMMA, tackles the problem by nding a hyperplane with a minimal norm under two linear constraints. The rst constraint is presented so that the new classi er will classify well previous examples, while the second rule demands that the hyperplane will ....
....attempts to update the matrix M on each round regardless of whether there was a prediction error or not. We show below that the algorithm is ultraconservative and thus t is the zero vector if x t is correctly classi ed (and no update takes place) Following the trend set by Li and Long [13] and Gentile [10] we term our algorithm MIRA for Margin Infused Relaxed Algorithm. The algorithm is described in Fig. 3. Before investigating the properties of the algorithm, we rewrite the optimization problem that MIRA solves on each round in a more convenient form. Omitting the example index t ....
[Article contains additional citation context not shown here]
Yi Li and Phil M. Long. The relaxed online maximum margin algorithm. In Advances in Neural Information Processing Systems 13, 1999.
Journal of Machine Learning Research 7 (2006) 1205--1230 .. - For Linear.. (Correct)
No context found.
Y. Li and P. Long. The relaxed online maximum margin algorithm. Machine Learning, 46:361--387, 2002.
Journal of Machine Learning Research 7 (2006) 551--585.. - Koby Crammer Crammer (Correct)
No context found.
Y. Li and P. M. Long. The relaxed online maximum margin algorithm. Machine Learning, 46(1--3): 361--387, 2002.
Journal of Machine Learning Research 6 (2005) 1579--1619.. - With Online And (2005) (Correct)
No context found.
Y. Li and P. Long. The relaxed online maximum margin algorithm. Machine Learning, 46:361--387, 2002.
Online Policy Adaptation for Ensemble Classifiers - Dimitrakakis, Bengio (2004) (Correct)
No context found.
Yi Li and Philip M. Long. The relaxed online maximum margin algorithm. Machine Learning, 46(1/3):361, 2002.
Online Adaptive Policies for Ensemble Classifiers - Dimitrakakis, Bengio (2004) (Correct)
No context found.
Y. Li, P. M. Long, The relaxed online maximum margin algorithm, Machine Learning 46 (1/3) (2002) 361. A Classification Margin The margin distribution for the two class case can be defined as:
Unknown - Pe Cep Ua (Correct)
No context found.
Yi Li and Philip M. Long. The relaxed online maximum margin algorithm. Machine Learning, 46(1/3):361, 2002.
Online (and Offline) on an Even Tighter Budget - Jason Weston America (Correct)
No context found.
Li, Y. and Long, P. (2002). The Relaxed Online Maximum Margin Algorithm. Machine Learning, 46:361--387.
Online Learning with Kernels - Kivinen, Smola, Williamson (2002) (5 citations) (Correct)
No context found.
Y. Li and P. M. Long, "The relaxed online maximum margin algorithm," Machine Learning, vol. 46, no. 1, pp. 361--387, Jan. 2002.
Large Margin Classification for Moving Targets - Kivinen, Smola, Williamson (2001) (Correct)
No context found.
Y. Li and P. M. Long. The relaxed online maximum margin algorithm. Machine , 46(1):361--387, January 2002.
Online Learning with Kernels - Kivinen, Smola, Williamson (2002) (5 citations) (Correct)
No context found.
Y. Li and P.M. Long. The relaxed online maximum margin algorithm. In S. A. Solla, T. K. Leen, and K.-R. M uller, editors, Advances in Neural Information Processing Systems 12, pages 498--504. MIT Press, 1999.
Online Classification on a Budget - Crammer, Kandola, Singer (2003) (2 citations) (Correct)
No context found.
Y. Li and P. M. Long. The relaxed online maximum margin algorithm. Machine Learning, 46(1--3):361--387, 2002.
Online Passive-Aggressive Algorithms - Crammer, Dekel, Shalev-Shwartz.. (2003) (Correct)
No context found.
Y. Li and P. M. Long. The relaxed online maximum margin algorithm. Machine Learning, 46(1--3):361--387, 2002.
Online Policy Adaption for Ensemble Algorithms - Dimitrakakis, Bengio (2002) (Correct)
No context found.
Yi Li and Philip M. Long. The relaxed online maximum margin algorithm. Machine Learning, 46(1/3):361, 2002.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC