{"id":5961,"date":"2026-05-27T19:42:43","date_gmt":"2026-05-27T19:42:43","guid":{"rendered":"https:\/\/quantil.co\/?post_type=blog&#038;p=5961"},"modified":"2026-05-27T19:45:15","modified_gmt":"2026-05-27T19:45:15","slug":"cuando-equivocarse-no-importa-repensando-como-entrenamos-modelos-que-toman-decisiones","status":"publish","type":"blog","link":"https:\/\/quantil.co\/en\/blog\/cuando-equivocarse-no-importa-repensando-como-entrenamos-modelos-que-toman-decisiones\/","title":{"rendered":"Cuando equivocarse no importa: repensando c\u00f3mo entrenamos modelos que toman decisiones"},"excerpt":{"rendered":"<p>La forma est\u00e1ndar de evaluar modelos predictivos es dominada por una idea simple: si baja el error de predicci\u00f3n, el modelo es mejor. M\u00e9tricas como el MSE o el accuracy se han convertido en el est\u00e1ndar en la mayor\u00eda de pipelines industriales &#8230;<\/p>\n","protected":false},"featured_media":5963,"template":"","format":"standard","blog-categories":[28],"blog-tags":[83],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/quantil.co\/en\/wp-json\/wp\/v2\/blog\/5961"}],"collection":[{"href":"https:\/\/quantil.co\/en\/wp-json\/wp\/v2\/blog"}],"about":[{"href":"https:\/\/quantil.co\/en\/wp-json\/wp\/v2\/types\/blog"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/quantil.co\/en\/wp-json\/wp\/v2\/media\/5963"}],"wp:attachment":[{"href":"https:\/\/quantil.co\/en\/wp-json\/wp\/v2\/media?parent=5961"}],"wp:term":[{"taxonomy":"blog-categories","embeddable":true,"href":"https:\/\/quantil.co\/en\/wp-json\/wp\/v2\/blog-categories?post=5961"},{"taxonomy":"blog-tags","embeddable":true,"href":"https:\/\/quantil.co\/en\/wp-json\/wp\/v2\/blog-tags?post=5961"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}