Online Passive-Aggressive Algorithms

Online Passive-Aggressive Algorithms. Crammer, K., Dekel, O., Keshet, J., Shalev-Shwartz, S., & Singer, Y. Journal of Machine Learning Research, 2006.
abstract bibtex

We present a family of margin based online learning algorithms for various prediction tasks. In particular we derive and analyze algorithms for binary and multiclass categorization, regression, uniclass prediction and sequence prediction. The update steps of our different algorithms are all based on analytical solutions to simple constrained optimization problems. This uniﬁed view allows us to prove worst-case loss bounds for the different algorithms and for the various decision problems based on a single lemma. Our bounds on the cumulative loss of the algorithms are relative to the smallest loss that can be attained by any ﬁxed hypothesis, and as such are applicable to both realizable and unrealizable settings. We demonstrate some of the merits of the proposed algorithms in a series of experiments with synthetic and real data sets.

@article{crammer_online_2006,
	title = {Online {Passive}-{Aggressive} {Algorithms}},
	abstract = {We present a family of margin based online learning algorithms for various prediction tasks. In particular we derive and analyze algorithms for binary and multiclass categorization, regression, uniclass prediction and sequence prediction. The update steps of our different algorithms are all based on analytical solutions to simple constrained optimization problems. This uniﬁed view allows us to prove worst-case loss bounds for the different algorithms and for the various decision problems based on a single lemma. Our bounds on the cumulative loss of the algorithms are relative to the smallest loss that can be attained by any ﬁxed hypothesis, and as such are applicable to both realizable and unrealizable settings. We demonstrate some of the merits of the proposed algorithms in a series of experiments with synthetic and real data sets.},
	language = {en},
	journal = {Journal of Machine Learning Research},
	author = {Crammer, Koby and Dekel, Ofer and Keshet, Joseph and Shalev-Shwartz, Shai and Singer, Yoram},
	year = {2006},
	keywords = {⛔ No DOI found},
	pages = {35},
}

Downloads: 0

{"_id":"4PxkfwfDqAyWzPykE","bibbaseid":"crammer-dekel-keshet-shalevshwartz-singer-onlinepassiveaggressivealgorithms-2006","author_short":["Crammer, K.","Dekel, O.","Keshet, J.","Shalev-Shwartz, S.","Singer, Y."],"bibdata":{"bibtype":"article","type":"article","title":"Online Passive-Aggressive Algorithms","abstract":"We present a family of margin based online learning algorithms for various prediction tasks. In particular we derive and analyze algorithms for binary and multiclass categorization, regression, uniclass prediction and sequence prediction. The update steps of our different algorithms are all based on analytical solutions to simple constrained optimization problems. This uniﬁed view allows us to prove worst-case loss bounds for the different algorithms and for the various decision problems based on a single lemma. Our bounds on the cumulative loss of the algorithms are relative to the smallest loss that can be attained by any ﬁxed hypothesis, and as such are applicable to both realizable and unrealizable settings. We demonstrate some of the merits of the proposed algorithms in a series of experiments with synthetic and real data sets.","language":"en","journal":"Journal of Machine Learning Research","author":[{"propositions":[],"lastnames":["Crammer"],"firstnames":["Koby"],"suffixes":[]},{"propositions":[],"lastnames":["Dekel"],"firstnames":["Ofer"],"suffixes":[]},{"propositions":[],"lastnames":["Keshet"],"firstnames":["Joseph"],"suffixes":[]},{"propositions":[],"lastnames":["Shalev-Shwartz"],"firstnames":["Shai"],"suffixes":[]},{"propositions":[],"lastnames":["Singer"],"firstnames":["Yoram"],"suffixes":[]}],"year":"2006","keywords":"⛔ No DOI found","pages":"35","bibtex":"@article{crammer_online_2006,\n\ttitle = {Online {Passive}-{Aggressive} {Algorithms}},\n\tabstract = {We present a family of margin based online learning algorithms for various prediction tasks. In particular we derive and analyze algorithms for binary and multiclass categorization, regression, uniclass prediction and sequence prediction. The update steps of our different algorithms are all based on analytical solutions to simple constrained optimization problems. This uniﬁed view allows us to prove worst-case loss bounds for the different algorithms and for the various decision problems based on a single lemma. Our bounds on the cumulative loss of the algorithms are relative to the smallest loss that can be attained by any ﬁxed hypothesis, and as such are applicable to both realizable and unrealizable settings. We demonstrate some of the merits of the proposed algorithms in a series of experiments with synthetic and real data sets.},\n\tlanguage = {en},\n\tjournal = {Journal of Machine Learning Research},\n\tauthor = {Crammer, Koby and Dekel, Ofer and Keshet, Joseph and Shalev-Shwartz, Shai and Singer, Yoram},\n\tyear = {2006},\n\tkeywords = {⛔ No DOI found},\n\tpages = {35},\n}\n\n\n\n","author_short":["Crammer, K.","Dekel, O.","Keshet, J.","Shalev-Shwartz, S.","Singer, Y."],"key":"crammer_online_2006","id":"crammer_online_2006","bibbaseid":"crammer-dekel-keshet-shalevshwartz-singer-onlinepassiveaggressivealgorithms-2006","role":"author","urls":{},"keyword":["⛔ No DOI found"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/fsimonetta","dataSources":["iwKepCrWBps7ojhDx","pzyFFGWvxG2bs63zP"],"keywords":["⛔ no doi found"],"search_terms":["online","passive","aggressive","algorithms","crammer","dekel","keshet","shalev-shwartz","singer"],"title":"Online Passive-Aggressive Algorithms","year":2006}