<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Machine Learning Optimization on C.CUI's Log</title><link>https://cuicaihao.github.io/tags/machine-learning-optimization/</link><description>Recent content in Machine Learning Optimization on C.CUI's Log</description><generator>Hugo</generator><language>en-AU</language><lastBuildDate>Tue, 25 Aug 2026 07:00:00 +1000</lastBuildDate><atom:link href="https://cuicaihao.github.io/tags/machine-learning-optimization/index.xml" rel="self" type="application/rss+xml"/><item><title>How AI Foundation Models Address Scaling Bottlenecks: Mathematical Structures, Algorithm Design, and Systems Engineering</title><link>https://cuicaihao.github.io/posts/2026-08-25-mathematical-and-algorithmic-ideas-that-changed-ai-over-the-last-decade-from-scaling-bottlenecks-to-foundation-models/</link><pubDate>Tue, 25 Aug 2026 07:00:00 +1000</pubDate><guid>https://cuicaihao.github.io/posts/2026-08-25-mathematical-and-algorithmic-ideas-that-changed-ai-over-the-last-decade-from-scaling-bottlenecks-to-foundation-models/</guid><description>Over the past decade, foundation models have advanced through more than additional data, parameters, and compute. Researchers have also identified useful structure in hard problems and redesigned parameter representations, computation order, resource allocation, generative paths, training objectives, and runtime state so that previously intractable problems can be solved at scale.</description></item></channel></rss>