# 第6章メモ ## 6.1 双対表現 ### P.3について $$ J(\mathbf{w})=\frac{1}{2} \sum_{n=1}^{N}\left\{\mathbf{w}^{\mathrm T} \boldsymbol{\phi}\left(\mathbf{x}_{n}\right)-t_{n}\right\}^{2}+\frac{\lambda}{2} \mathbf{w}^{\mathrm T} \mathbf{w} $$ $\displaystyle \frac{\partial J}{\partial \mathbf{w}}=0$とすると $$ \frac{\partial J}{\partial \mathbf{w}}=\sum_{n=1}^{N}\left\{\mathbf{w}^{\mathrm{T}} \boldsymbol{\phi}\left(\mathbf{x}_{n}\right)-t_{n}\right\} \boldsymbol{\phi}\left(\mathbf{x}_{n}\right)+\frac{\lambda}{2} \cdot 2 \mathbf{w}=0 $$ $$ \begin{aligned} \mathbf{w} &=-\frac{1}{\lambda} \sum_{n=1}^{N}\left\{\mathbf{w}^{\mathrm T} \boldsymbol{\phi}\left(\mathbf{x}_{n}\right)-t_{n}\right\} \boldsymbol{\phi}\left(\mathbf{x}_{n}\right) \\ &= \sum_{n=1}^{N}\underbrace{\left( -\frac{1}{\lambda}\left\{\mathbf{w}^{\mathrm T} \boldsymbol{\phi}\left(\mathbf{x}_{n}\right)-t_{n}\right\}\right)}_{a_{n}}\boldsymbol{\phi}\left(\mathbf{x}_{n}\right) \\ &=\mathbf{\Phi}^{\mathrm T}\mathbf{a} \end{aligned} $$ とする。ここで$\mathbf{\Phi}$は$n$番目の行が$\boldsymbol{\phi}(\mathbf{x}_n)^{\mathrm T}$で表される計画行列である。 $$ \mathbf{\Phi}^{\mathrm T}\mathbf{a}=\left(\boldsymbol{\phi}\left(\mathbf{x}_1\right), \boldsymbol{\phi}\left(\mathbf{x}_2\right) \cdots \boldsymbol{\phi}\left(\mathbf{x}_n\right)\right)\left(\begin{array}{c} a_{1} \\ a_{2} \\ \vdots \\ a_{n} \end{array}\right) $$
×
Sign in
Email
Password
Forgot password
or
By clicking below, you agree to our
terms of service
.
Sign in via Facebook
Sign in via Twitter
Sign in via GitHub
Sign in via Dropbox
Sign in with Wallet
Wallet (
)
Connect another wallet
New to HackMD?
Sign up