AdamNi_NintyNine的博客

看最优化时发现许多围绕PSD、Eigenvectors等概念的定理不太熟悉，于是想着把线代基础重新过一遍。打算过两套材料：MIT Gilbert Strang的线代课，老经典了，很基础。可惜以前没看过，借此机会膜拜一下，再锻炼锻炼直觉Linear Algebra by Kenneth Hoffman & Ray Kunze。重点在后半部分inner product spaces, 可能要花两周的时间重新学一下。笔记只记能增进直觉的核心理解。以前实分析的老师说，他觉得反证法不算证明，因为很多

Disclaimer: This is a study note of the Fall 2018 Convex Optimization course offered at CMU. The note is compiled to document key understandings of concepts, which I think greatly aids the memorization and recalling process. While the note will inevitably

1. Close-to-close公式: σ=Σi=1i=N(ri−μ)2N\sigma = \sqrt{\frac{\Sigma_{i = 1}^{i = N} (r_i - \mu)^2}{N}}σ=NΣi=1i=N​(ri​−μ)2​​ 再年化优点: 简单易懂缺点: 没隔夜跳空，没日内信息，容易低估。2. Parkinson 1980. aka High low HV公式: σ=Σi=1i=N(lnhili)24Nln2\sigma = \sqrt{\frac{\Sigma_{i = 1}

Variable Passing in Python2021.12.29 To be updated after machine structure is learntSource: A very good graphical explanationWe think of varible x = [1, 2] using a box-pointer diagram, where x points to a box containing a pointer to the list object [1,