{"id":4700,"date":"2025-11-12T22:45:33","date_gmt":"2025-11-12T22:45:33","guid":{"rendered":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?page_id=4700"},"modified":"2025-11-12T22:45:33","modified_gmt":"2025-11-12T22:45:33","slug":"confidence-calibration-for-weighted-voting-in-rf-ensembles-2","status":"publish","type":"page","link":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/?page_id=4700","title":{"rendered":"Open-Set Handling in RF Ensembles: Thresholding, Abstention, and OSCR Analysis"},"content":{"rendered":"\n<div data-wp-interactive=\"core\/file\" class=\"wp-block-file\"><object data-wp-bind--hidden=\"!state.hasPdfPreview\" hidden class=\"wp-block-file__embed\" data=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Open-Set-Handling-in-RF-Ensembles-Thresholding-Abstention-and-OSCR-Analysis-bgilbert1984-Rev-2.pdf\" type=\"application\/pdf\" style=\"width:100%;height:600px\" aria-label=\"Embed of Open-Set Handling in RF Ensembles Thresholding, Abstention, and OSCR Analysis bgilbert1984 Rev 2.\"><\/object><a id=\"wp-block-file--media-e6085c6d-bf7b-4493-b519-c95a477dc5e9\" href=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Open-Set-Handling-in-RF-Ensembles-Thresholding-Abstention-and-OSCR-Analysis-bgilbert1984-Rev-2.pdf\">Open-Set Handling in RF Ensembles Thresholding, Abstention, and OSCR Analysis bgilbert1984 Rev 2<\/a><a href=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Open-Set-Handling-in-RF-Ensembles-Thresholding-Abstention-and-OSCR-Analysis-bgilbert1984-Rev-2.pdf\" class=\"wp-block-file__button wp-element-button\" download aria-describedby=\"wp-block-file--media-e6085c6d-bf7b-4493-b519-c95a477dc5e9\">Download<\/a><\/div>\n\n\n\n<p class=\"wp-block-paragraph\">Operational RF classifiers encounter signals outside<br>the closed-set label taxonomy. We treat Unknown as a first-class<br>outcome via score thresholding on logits-derived confidence (maxprobability), predictive entropy, and energy score. We evaluate<br>with OSCR curves (Correct Classification Rate vs. Unknown<br>False-Positive Rate) and AU-PR for unknown detection. With<br>lightweight hooks in the ensemble probability path, we realize<br>robust abstention with &lt;0.1 ms overhead and maintain utility<br>(accuracy\u00d7coverage) across SNR bins.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Open-set recognition (OSR) addresses the mismatch between training labels and real-world observations. In RFML<br>pipelines, abstention is often treated as a failure; we instead<br>elevate Unknown to a policy outcome with measurable tradeoffs. We analyze simple gates on: (i) max softmax, (ii)<br>predictive entropy, and (iii) logit-energy, and report OSCR [1],<br>[2] and AU-PR(Unknown).<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Related Papers<\/p>\n\n\n\n<div data-wp-interactive=\"core\/file\" class=\"wp-block-file\"><object data-wp-bind--hidden=\"!state.hasPdfPreview\" hidden class=\"wp-block-file__embed\" data=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Confidence-Calibration-for-Weighted-Voting-in-RF-Ensembles-ATL-Features-bgilbert1984-Rev2.pdf\" type=\"application\/pdf\" style=\"width:100%;height:600px\" aria-label=\"Embed of Confidence Calibration for Weighted Voting in RF Ensembles ATL Features bgilbert1984 Rev2.\"><\/object><a id=\"wp-block-file--media-ba0fb7a0-7b8a-4d77-a6a6-faec34dab8db\" href=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Confidence-Calibration-for-Weighted-Voting-in-RF-Ensembles-ATL-Features-bgilbert1984-Rev2.pdf\">Confidence Calibration for Weighted Voting in RF Ensembles ATL Features bgilbert1984 Rev2<\/a><a href=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Confidence-Calibration-for-Weighted-Voting-in-RF-Ensembles-ATL-Features-bgilbert1984-Rev2.pdf\" class=\"wp-block-file__button wp-element-button\" download aria-describedby=\"wp-block-file--media-ba0fb7a0-7b8a-4d77-a6a6-faec34dab8db\">Download<\/a><\/div>\n\n\n\n<div data-wp-interactive=\"core\/file\" class=\"wp-block-file\"><object data-wp-bind--hidden=\"!state.hasPdfPreview\" hidden class=\"wp-block-file__embed\" data=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Resampling-Effects-FFT-256-Seq-128-Rev-3-bgilbert1984.pdf\" type=\"application\/pdf\" style=\"width:100%;height:600px\" aria-label=\"Embed of Resampling Effects FFT 256 Seq 128 Rev 3 bgilbert1984.\"><\/object><a id=\"wp-block-file--media-5922f05c-2304-484a-8a6f-2deb5544f71c\" href=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Resampling-Effects-FFT-256-Seq-128-Rev-3-bgilbert1984.pdf\">Resampling Effects FFT 256 Seq 128 Rev 3 bgilbert1984<\/a><a href=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Resampling-Effects-FFT-256-Seq-128-Rev-3-bgilbert1984.pdf\" class=\"wp-block-file__button wp-element-button\" download aria-describedby=\"wp-block-file--media-5922f05c-2304-484a-8a6f-2deb5544f71c\">Download<\/a><\/div>\n","protected":false},"excerpt":{"rendered":"<p>Operational RF classifiers encounter signals outsidethe closed-set label taxonomy. We treat Unknown as a first-classoutcome via score thresholding on logits-derived confidence (maxprobability), predictive entropy, and energy score. We evaluatewith OSCR curves (Correct Classification Rate vs. UnknownFalse-Positive Rate) and AU-PR for unknown detection. Withlightweight hooks in the ensemble probability path, we realizerobust abstention with &lt;0.1 ms&hellip;&nbsp;<\/p>\n","protected":false},"author":2,"featured_media":2900,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"neve_meta_sidebar":"","neve_meta_container":"","neve_meta_enable_content_width":"","neve_meta_content_width":0,"neve_meta_title_alignment":"","neve_meta_author_avatar":"","neve_post_elements_order":"","neve_meta_disable_header":"","neve_meta_disable_footer":"","neve_meta_disable_title":"","footnotes":""},"class_list":["post-4700","page","type-page","status-publish","has-post-thumbnail","hentry"],"_links":{"self":[{"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=\/wp\/v2\/pages\/4700","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=4700"}],"version-history":[{"count":0,"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=\/wp\/v2\/pages\/4700\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=\/wp\/v2\/media\/2900"}],"wp:attachment":[{"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4700"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}