Matthieu R Bloch
May 11, 2020
\[ \hat{p} = \frac{\text{# head}}{N} \]
It is possible that \(\hat{p}\) is completely off but it is not probable
An unknown function \(f:\calX\to\calY:\bfx\mapsto y=f(\bfx)\) to learn
A set of hypotheses \(\calH\) as to what the function could be
An algorithm \(\texttt{ALG}\) to find the best \(h\in\calH\) that explains \(f\)
A set of hypotheses \(\calH\) as to what the function could be
A set of hypotheses \(\calH\) as to what the function could be
A loss function \(\ell:\calY\times\calY\rightarrow\bbR^+\) capturing the “cost” of prediction
An algorithm \(\texttt{ALG}\) to find the best \(h\in\calH\) that explains \(f\)
Quick demo: nearest neighbor classification