Essec\Faculty\Model\Contribution {#2216
#_index: "academ_contributions"
#_id: "15416"
#_source: array:26 [
"id" => "15416"
"slug" => "15416-learning-with-a-linear-loss-function-excess-risk-and-estimation-bounds-for-erm-minmax-mom-and-their-regularized-versions-with-applications-to-robustness-in-sparse-pca"
"yearMonth" => "2025-01"
"year" => "2025"
"title" => "Learning with a linear loss function: excess risk and estimation bounds for ERM, minmax MOM and their regularized versions with applications to robustness in sparse PCA"
"description" => "LECUE, G. et NEIRAC, L. (2025). Learning with a linear loss function: excess risk and estimation bounds for ERM, minmax MOM and their regularized versions with applications to robustness in sparse PCA. <i>Journal of Machine Learning Research</i>, 29(399), pp. 1-90."
"authors" => array:2 [
0 => array:3 [
"name" => "LECUE Guillaume"
"bid" => "B00806953"
"slug" => "lecue-guillaume"
]
1 => array:1 [
"name" => "NEIRAC Lucie"
]
]
"ouvrage" => ""
"keywords" => array:6 [
0 => "SDP relaxation"
1 => "empirical processes"
2 => "robustness"
3 => "heavy-tailed"
4 => "adversarial contamination"
5 => "high-dimensional statistics"
]
"updatedAt" => "2025-03-03 14:01:04"
"publicationUrl" => "https://www.jmlr.org/papers/v25/23-1405.html"
"publicationInfo" => array:3 [
"pages" => "1-90"
"volume" => "29"
"number" => "399"
]
"type" => array:2 [
"fr" => "Articles"
"en" => "Journal articles"
]
"support_type" => array:2 [
"fr" => "Revue scientifique"
"en" => "Scientific journal"
]
"countries" => array:2 [
"fr" => "États-Unis"
"en" => "United States of America"
]
"abstract" => array:2 [
"fr" => "Motivated by several examples, we consider a general framework of learning with linear loss functions. In this context, we provide excess risk and estimation bounds that hold with large probability for four estimators: ERM, minmax MOM and their regularized versions. These general bounds are applied for the problem of robustness in sparse PCA. In particular, we improve the state of the art result for this this problems, obtain results under weak moment assumptions as well as for adversarial contaminated data."
"en" => "Motivated by several examples, we consider a general framework of learning with linear loss functions. In this context, we provide excess risk and estimation bounds that hold with large probability for four estimators: ERM, minmax MOM and their regularized versions. These general bounds are applied for the problem of robustness in sparse PCA. In particular, we improve the state of the art result for this this problems, obtain results under weak moment assumptions as well as for adversarial contaminated data."
]
"authors_fields" => array:2 [
"fr" => "Systèmes d'Information, Data Analytics et Opérations"
"en" => "Information Systems, Data Analytics and Operations"
]
"indexedAt" => "2025-07-18T03:21:56.000Z"
"docTitle" => "Learning with a linear loss function: excess risk and estimation bounds for ERM, minmax MOM and their regularized versions with applications to robustness in sparse PCA"
"docSurtitle" => "Journal articles"
"authorNames" => "<a href="/cv/lecue-guillaume">LECUE Guillaume</a>, NEIRAC Lucie"
"docDescription" => "<span class="document-property-authors">LECUE Guillaume, NEIRAC Lucie</span><br><span class="document-property-authors_fields">Information Systems, Data Analytics and Operations</span> | <span class="document-property-year">2025</span>"
"keywordList" => "<a href="#">SDP relaxation</a>, <a href="#">empirical processes</a>, <a href="#">robustness</a>, <a href="#">heavy-tailed</a>, <a href="#">adversarial contamination</a>, <a href="#">high-dimensional statistics</a>"
"docPreview" => "<b>Learning with a linear loss function: excess risk and estimation bounds for ERM, minmax MOM and their regularized versions with applications to robustness in sparse PCA</b><br><span>2025-01 | Journal articles </span>"
"docType" => "research"
"publicationLink" => "<a href="https://www.jmlr.org/papers/v25/23-1405.html" target="_blank">Learning with a linear loss function: excess risk and estimation bounds for ERM, minmax MOM and their regularized versions with applications to robustness in sparse PCA</a>"
]
+lang: "en"
+"_type": "_doc"
+"_score": 8.7374525
+"parent": null
}