Machine Learning Garden

Powered by 🌱Roam Garden

However, one numerical problem here is that we are multiply together many numbers less than one.

To solve the problem, we can use log\log to convert multiplication into addition.

logp(D)=ilogp(xi)+logpθ(yixi)=ilogpθ(yixi)+ const \log p(\mathcal{D})=\sum_{i} \log p\left(x_{i}\right)+\log p_{\theta}\left(y_{i} \mid x_{i}\right) =\sum_{i} \log p_{\theta}\left(y_{i} \mid x_{i}\right)+\text { const }

θargmaxθilogpθ(yixi)\theta^{\star} \leftarrow \arg \max _{\theta} \sum_{i} \log p_{\theta}\left(y_{i} \mid x_{i}\right)

Referenced in

Maximum Likelihood Estimation (MLE)