{"skill":{"slug":"lowdim-hidim","displayName":"Low-Dim Solve High-Dim","summary":"用低维方法解决高维复杂系统问题 - 降维建模、低维表征、子空间分解、稀疏约束。突破维度灾难，降低计算成本。应用于随机控制、强化学习、信号处理、AI推理等。","tags":{"dimension-reduction":"1.0.0","latest":"1.0.0","machine-learning":"1.0.0","mathematics":"1.0.0","optimization":"1.0.0","stochastic-control":"1.0.0"},"stats":{"comments":0,"downloads":68,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":1},"createdAt":1776404415815,"updatedAt":1776404812248},"latestVersion":{"version":"1.0.0","createdAt":1776404415815,"changelog":"- Initial release of lowdim-hidim (v1.0.0) — low-dimensional methods for high-dimensional complex systems.\n- Includes explanations of \"curse of dimensionality\" and motivation for dimension reduction.\n- Summarizes core techniques: PCA, t-SNE, UMAP, autoencoders, subspace learning, sparse representation, low-rank approximations, manifold learning, mean-field approximations, and LQG reduction.\n- Details typical applications: stochastic control, reinforcement learning, signal processing, large AI model inference, industrial control.\n- Provides method selection guide and Q&A for practical usage.\n- Features an easy-to-understand science communication section for beginners.","license":"MIT-0"},"metadata":null,"owner":{"handle":"smseow001","userId":"s17a72fh5mgpvvs3dm2ax0yq8984pnqc","displayName":"SMS","image":"https://avatars.githubusercontent.com/u/33817483?v=4"},"moderation":null}