{"feed_url":"https://mathhub.eggtea.org/feed.json","home_page_url":"https://mathhub.eggtea.org","items":[{"content_text":"For strongly convex smooth objectives and deterministic historical gradient-error weights, the notes derive convergence and a PR CLT whose effective covariance depends on cumulative weights and innovation cross-covariance. Running averages and fixed c/t gains illustrate different single-step and downstream objectives. Within normalized deterministic scalar historical linear estimators, a sharp cumulative variance bound is attained by a polynomial-gain recursion under the stated exponent conditions.","date_published":"2026-10-17T09:22:33+00:00","id":"https://mathhub.eggtea.org/works/sgd-with-vanishing-correlated-gradient-errors-dc3bf255/v/1","title":"SGD with Vanishing Correlated Gradient Errors (technique notes)","url":"https://mathhub.eggtea.org/works/sgd-with-vanishing-correlated-gradient-errors-dc3bf255/v/1"},{"content_text":"A twenty-one-case research draft on rational Hodge classes of complex abelian sixfolds. The overall sixfold conclusion is conditional on the manuscript-specific split-eightfold Weil, CM, complete Kuga\u2013Satake and auxiliary spin-tensor algebraicity assertions. Individual unconditional strata and all-powers results have the narrower scopes recorded in the manuscript and theorem registry; no unconditional resolution is claimed.","date_published":"2026-10-17T09:22:31+00:00","id":"https://mathhub.eggtea.org/works/hodge-classes-on-abelian-sixfolds-constructio-82bdae05/v/1","title":"Hodge Classes on Abelian Sixfolds: Constructions and a Conditional Proof","url":"https://mathhub.eggtea.org/works/hodge-classes-on-abelian-sixfolds-constructio-82bdae05/v/1"},{"content_text":"For entire input scores under exponential derivative envelopes, amplitude control and suitable sampling moments, an imaginary Gaussian perturbation yields an unbiased clean-gradient oracle. The manuscript derives moment bounds, convergence rates and last-iterate and averaged CLTs, with a distinct averaging covariance at the inverse-time endpoint.","date_published":"2026-10-17T09:22:24+00:00","id":"https://mathhub.eggtea.org/works/complex-corrected-stochastic-gradient-descent-3ac3e220/v/1","title":"Complex-Corrected Stochastic Gradient Descent for Noisy Observations","url":"https://mathhub.eggtea.org/works/complex-corrected-stochastic-gradient-descent-3ac3e220/v/1"}],"title":"MathHub","version":"https://jsonfeed.org/version/1.1"}