{"id":4529,"date":"2025-11-08T17:42:31","date_gmt":"2025-11-08T17:42:31","guid":{"rendered":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?page_id=4529"},"modified":"2025-11-08T17:42:31","modified_gmt":"2025-11-08T17:42:31","slug":"majority-vs-weighted-vs-stacked-voting-in-rf-modulation-ensembles","status":"publish","type":"page","link":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/?page_id=4529","title":{"rendered":"Majority vs Weighted vs Stacked Voting in RF Modulation Ensembles"},"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\/Majority-vs-Weighted-vs-Stacked-Voting-in-RF-Modulation-Ensembles-bgilbert1984.pdf\" type=\"application\/pdf\" style=\"width:100%;height:600px\" aria-label=\"Embed of Majority vs Weighted vs Stacked Voting in RF Modulation Ensembles bgilbert1984.\"><\/object><a id=\"wp-block-file--media-6113ba84-bafd-4c5b-944c-c88e70ce2aad\" href=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Majority-vs-Weighted-vs-Stacked-Voting-in-RF-Modulation-Ensembles-bgilbert1984.pdf\">Majority vs Weighted vs Stacked Voting in RF Modulation Ensembles bgilbert1984<\/a><a href=\"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/wp-content\/uploads\/2025\/11\/Majority-vs-Weighted-vs-Stacked-Voting-in-RF-Modulation-Ensembles-bgilbert1984.pdf\" class=\"wp-block-file__button wp-element-button\" download aria-describedby=\"wp-block-file--media-6113ba84-bafd-4c5b-944c-c88e70ce2aad\">Download<\/a><\/div>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Performance Metrics.** <strong>Key Findings<\/strong> includes a bullet point stating, &#8220;Stacked voting showed the best calibration (lowest ECE) at $K=3$ .&#8221; <strong>Accuracy vs. Models<\/strong> shows a simplified line chart illustrating that all three methods (majority, weighted, stacked) reach 1.0 accuracy at $K=3$ models, with stacked showing higher intermediate accuracy at $K=2$<sup>1<\/sup>. <strong>Performance Metrics<\/strong> is a table showing: <strong>Metric<\/strong>: TTFB (p50) at $K=4$, <strong>Majority<\/strong>: 3.2 ms, <strong>Weighted<\/strong>: 3.2 ms, <strong>Stacked<\/strong>: 3.4 ms<sup>2<\/sup><sup>2<\/sup><sup>2<\/sup>. Another row shows: <strong>Metric<\/strong>: ECE at $K=3$, <strong>Majority<\/strong>: 0.654, <strong>Weighted<\/strong>: 0.654, <strong>Stacked<\/strong>: 0.333<sup>3<\/sup><sup>3<\/sup><sup>3<\/sup>. A final note states, &#8220;Weighted voting typically dominates majority when confidences are calibrated; stacked can surpass both given diverse base-model errors and sufficient meta-data<sup>4<\/sup>.&#8221;]<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h3 class=\"wp-block-heading\">\ud83d\uddbc\ufe0f RF Modulation Ensembles: Voting Strategy Comparison<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">This image summarizes the core results from the paper &#8220;Majority vs Weighted vs Stacked Voting in RF Modulation Ensembles&#8221;<sup>5<\/sup><sup>5<\/sup><sup>5<\/sup><sup>5<\/sup>.<\/p>\n\n\n\n<h4 class=\"wp-block-heading\"><strong>1. Key Findings<\/strong><\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Accuracy:<\/strong> All three methods (majority, weighted, and stacked) achieved <strong>1.000 accuracy<\/strong> at <strong>$K=3$ models<\/strong><sup>6<\/sup>.<\/li>\n\n\n\n<li><strong>Calibration:<\/strong> <strong>Stacked voting<\/strong> demonstrated the best calibration, with the lowest Expected Calibration Error <strong>(ECE = 0.333)<\/strong> at $K=3$, compared to $0.654$ for majority and weighted<sup>7<\/sup><sup>7<\/sup><sup>7<\/sup>.<\/li>\n\n\n\n<li><strong>General Performance:<\/strong> <strong>Weighted voting<\/strong> is generally expected to outperform majority when confidences are calibrated, while <strong>stacked<\/strong> can exceed both if there are diverse base-model errors and enough meta-data<sup>8<\/sup>.<\/li>\n<\/ul>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h4 class=\"wp-block-heading\"><strong>2. Accuracy vs. Model Count ($K$)<\/strong><\/h4>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><td><strong># Models (K)<\/strong><\/td><td><strong>Majority<\/strong><\/td><td><strong>Weighted<\/strong><\/td><td><strong>Stacked<\/strong><\/td><\/tr><\/thead><tbody><tr><td><strong>1<\/strong><\/td><td>$\\approx 0.0$<\/td><td>$\\approx 0.0$<\/td><td>$\\approx 0.0$<\/td><\/tr><tr><td><strong>2<\/strong><\/td><td>$\\approx 0.0$<\/td><td>$\\approx 0.0$<\/td><td>$\\approx 0.35$<\/td><\/tr><tr><td><strong>3<\/strong><\/td><td><strong>1.000<\/strong><\/td><td><strong>1.000<\/strong><\/td><td><strong>1.000<\/strong><\/td><\/tr><tr><td><strong>4<\/strong><\/td><td>1.000<\/td><td>1.000<\/td><td>1.000<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\">Observation: Stacked voting showed a noticeable accuracy advantage at $K=2$ models before all three methods converged at $K=3$<sup>9<\/sup>.<\/p>\n<\/blockquote>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h4 class=\"wp-block-heading\"><strong>3. Performance Metrics at Max Model Count<\/strong><\/h4>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><td><strong>Metric<\/strong><\/td><td><strong>Majority<\/strong><\/td><td><strong>Weighted<\/strong><\/td><td><strong>Stacked<\/strong><\/td><\/tr><\/thead><tbody><tr><td><strong>Time-to-First-Byte (TTFB, p50) at $K=4$<\/strong><\/td><td>3.2 ms<\/td><td>3.2 ms<\/td><td>3.4 ms<\/td><\/tr><tr><td><strong>Expected Calibration Error (ECE) at $K=3$<\/strong><\/td><td>0.654<\/td><td>0.654<\/td><td><strong>0.333<\/strong><\/td><\/tr><tr><td><strong>Macro-F1 at $K=3$<\/strong><\/td><td>0.400<\/td><td>0.400<\/td><td>0.400<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">The data suggests that stacked voting is slightly slower in terms of median TTFB at $K=4$ models but provides significantly better calibration<sup>10<\/sup><sup>10<\/sup><sup>10<\/sup><sup>10<\/sup><sup>10<\/sup><sup>10<\/sup><sup>10<\/sup><sup>10<\/sup><sup>10<\/sup>.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p class=\"wp-block-paragraph\">Would you like to know more about the <strong>Stacked meta-learner<\/strong> used in the study or the different <strong>input types<\/strong> to the classifier?<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Performance Metrics.** Key Findings includes a bullet point stating, &#8220;Stacked voting showed the best calibration (lowest ECE) at $K=3$ .&#8221; Accuracy vs. Models shows a simplified line chart illustrating that all three methods (majority, weighted, stacked) reach 1.0 accuracy at $K=3$ models, with stacked showing higher intermediate accuracy at $K=2$1. Performance Metrics is a table&hellip;&nbsp;<\/p>\n","protected":false},"author":2,"featured_media":4531,"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-4529","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\/4529","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=4529"}],"version-history":[{"count":0,"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=\/wp\/v2\/pages\/4529\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=\/wp\/v2\/media\/4531"}],"wp:attachment":[{"href":"https:\/\/neurosphere-2.tail52f848.ts.net\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4529"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}