高斯回归权重空间过程
学习文献1 https://ww2.mathworks.cn/help/stats/gaussian-process-regression-models.html http://smellysheep.com/2018/07/%E9%AB%98%E6%96%AF%E8%BF%87%E7%A8%8B%E5%9B%9E%E5%BD%92/
( A + U C V ) − 1 = A − 1 − A − 1 U ( C − 1 + V A − 1 U ) − 1 V A − 1 (A+U C V)^{-1}=A^{-1}-A^{-1} U\left(C^{-1}+VA^{-1} U\right)^{-1} V A^{-1} (A+UCV)−1=A−1−A−1U(C−1+VA−1U)−1VA−1
方差： Φ ( x ∗ ) Σ p Φ ( x ∗ ) − Φ ( x ∗ ) T Σ p Φ ⊤ ( k + σ 2 I ) − 1 Φ Σ p Φ ( x ∗ ) \Phi\left(x^{*}\right) \Sigma_{p} \Phi\left(x^{*}\right)-\Phi\left(x^{*}\right)^{T} \Sigma p \Phi^{\top}\left(k+\sigma^{2} I\right)^{-1} \Phi \Sigma p \Phi\left(x^{*}\right) Φ(x∗)ΣpΦ(x∗)−Φ(x∗)TΣpΦ⊤(k+σ2I)−1ΦΣpΦ(x∗) k = Φ Σ p Φ T Φ ( x ∗ ) T Σ p Φ T Φ ( x ∗ ) Σ p Φ ( x ∗ ) Φ Σ p Φ ( x ∗ ) k=\Phi\Sigma_{p}\Phi^{T}\\ \Phi(x^*)^T\Sigma_p\Phi^T\\ \Phi(x^*)\Sigma_p\Phi(x^*)\\ \Phi\Sigma_p\Phi(x^*) k=ΦΣpΦTΦ(x∗)TΣpΦTΦ(x∗)ΣpΦ(x∗)ΦΣpΦ(x∗)
