{
    "author": null,
    "date_published": null,
    "dek": null,
    "direction": "ltr",
    "domain": "kylrth.com",
    "excerpt": "Posted on 2022-02-22 at 12:29:17 UTC-0500The big idea here is to use the geometric mean instead of the arithmetic mean across samples in the batch when computing the gradient for SGD. This overcomes&hellip;",
    "lead_image_url": null,
    "next_page_url": null,
    "rendered_pages": 1,
    "title": "deep-learning",
    "total_pages": 1,
    "url": "https://kylrth.com/tags/deep-learning/",
    "word_count": 1
}