{"id":9106,"date":"2025-12-15T07:02:31","date_gmt":"2025-12-15T07:02:31","guid":{"rendered":"https:\/\/mailitics.com\/index.php\/2025\/12\/15\/2512-11089\/"},"modified":"2025-12-15T07:02:31","modified_gmt":"2025-12-15T07:02:31","slug":"2512-11089","status":"publish","type":"post","link":"https:\/\/mailitics.com\/index.php\/2025\/12\/15\/2512-11089\/","title":{"rendered":"TPV: Parameter Perturbations Through the Lens of Test Prediction Variance"},"content":{"rendered":"

TPV: Parameter Perturbations Through the Lens of Test Prediction Variance
\n \t

\n
<\/BR>
\n
\n \t

\n
<\/BR><\/p>\n

arXiv:2512.11089v1 Announce Type: new
\nAbstract: We identify test prediction variance (TPV) — the first-order sensitivity of model outputs to parameter perturbations around a trained solution — as a unifying quantity that links several classical observations about generalization in deep networks. TPV is a fully label-free object whose trace form separates the geometry of the trained model from the specific perturbation mechanism, allowing a broad family of parameter perturbations like SGD noise, label noise, finite-precision noise, and other post-training perturbations to be analyzed under a single framework. Theoretically, we show that TPV estimated on the training set converges to its test-set value in the overparameterized limit, providing the first result that prediction variance under local parameter perturbations can be inferred from training inputs alone. Empirically, TPV exhibits a striking stability across datasets and architectures — including extremely narrow networks — and correlates well with clean test loss. Finally, we demonstrate that modeling pruning as a TPV perturbation yields a simple label-free importance measure that performs competitively with state-of-the-art pruning methods, illustrating the practical utility of TPV. Code available at github.com\/devansharpit\/TPV.<\/div>\n

\t

\n
<\/BR>
\n Devansh Arpit
\n \t

\n
<\/BR>
\nGo to original source<\/a>
\n \t

\n
<\/BR><\/p>\n","protected":false},"excerpt":{"rendered":"

TPV: Parameter Perturbations Through the Lens of Test Prediction Variance arXiv:2512.11089v1 Announce Type: new Abstract: We identify test prediction variance (TPV) — the first-order sensitivity of model outputs to parameter perturbations around a trained solution — as a unifying quantity that links several classical observations about generalization in deep networks. TPV is a fully label-free […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[62,113,112],"tags":[1452,3614,4429],"class_list":["post-9106","post","type-post","status-publish","format-standard","hentry","category-aimldsaimlds","category-cs-lg","category-stat-ml","tag-parameter","tag-perturbations","tag-tpv"],"_links":{"self":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/9106"}],"collection":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/comments?post=9106"}],"version-history":[{"count":0,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/9106\/revisions"}],"wp:attachment":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/media?parent=9106"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/categories?post=9106"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/tags?post=9106"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}