Commit 7f9cb92e by Ying-Qiu Zheng

### Update 2021JUL21.md

parent c19d572d
 ... @@ -23,10 +23,7 @@ In summary, in addition to finding the the hyper-parameters $\pi, \mu, \Sigma_{ ... @@ -23,10 +23,7 @@ In summary, in addition to finding the the hyper-parameters$\pi, \mu, \Sigma_{ 3. Initialise the transformation matrix $\mathbf{U}$ using Algorithm 3. 3. Initialise the transformation matrix $\mathbf{U}$ using Algorithm 3. 4. For iteration = $1, 2, ...$, do 4. For iteration = $1, 2, ...$, do - **E-step.** Evaluate the responsibilities using the current parameter values - **E-step.** Evaluate the responsibilities using the current parameter values - - $\gamma(y_{nk}) = \frac{\pi_{k}\mathcal{N}(\mathbf{x}^{L}_{n} | \mu_{k}, \Sigma_{k}^{L})\mathcal{N}(\mathbf{Ux}^{H}_{n} | \mu_{k}, \Sigma_{k}^{H})}{\sum_{j=1}^{K}\pi_{j}\mathcal{N}(\mathbf{x}^{L}_{n} | \mu_{k}, \Sigma_{k}^{L})\mathcal{N}(\mathbf{Ux}^{H}_{n} | \mu_{k}, \Sigma_{k}^{H})}$ math \gamma(y_{nk}) = \frac{\pi_{k}\mathcal{N}(\mathbf{x}^{L}_{n} | \mu_{k}, \Sigma_{k}^{L})\mathcal{N}(\mathbf{Ux}^{H}_{n} | \mu_{k}, \Sigma_{k}^{H})}{\sum_{j=1}^{K}\pi_{j}\mathcal{N}(\mathbf{x}^{L}_{n} | \mu_{k}, \Sigma_{k}^{L})\mathcal{N}(\mathbf{Ux}^{H}_{n} | \mu_{k}, \Sigma_{k}^{H})}  - **M-step.** Re-estimate the parameters using the current responsibilities by setting the derivatives to zero - **M-step.** Re-estimate the parameters using the current responsibilities by setting the derivatives to zero - $\mu_{k}^{\text{new}} = \frac{1}{N_{k}}((\Sigma^{H}_{k})^{-1} + (\Sigma^{L}_{k})^{-1} )^{-1}\sum_{n=1}^{N}\gamma(y_{nk})((\Sigma_{k}^{H})^{-1}\mathbf{Ux}^{H}_{n} + (\Sigma_{k}^{L})^{-1}\mathbf{x}_{n}^{L} )$ - $\mu_{k}^{\text{new}} = \frac{1}{N_{k}}((\Sigma^{H}_{k})^{-1} + (\Sigma^{L}_{k})^{-1} )^{-1}\sum_{n=1}^{N}\gamma(y_{nk})((\Sigma_{k}^{H})^{-1}\mathbf{Ux}^{H}_{n} + (\Sigma_{k}^{L})^{-1}\mathbf{x}_{n}^{L} )$ - $\Sigma_{k}^{L} = \frac{1}{N_{k}}\sum_{n=1}^{N}\gamma(y_{nk})(\mathbf{x}^{L}_{n} - \mathbf{\mu}_{k})(\mathbf{x}^{L}_{n} - \mathbf{\mu}_{k})^{T}$ - $\Sigma_{k}^{L} = \frac{1}{N_{k}}\sum_{n=1}^{N}\gamma(y_{nk})(\mathbf{x}^{L}_{n} - \mathbf{\mu}_{k})(\mathbf{x}^{L}_{n} - \mathbf{\mu}_{k})^{T}$ ... ...
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!