Commit a7e83b82 authored by Amelie Royer's avatar Amelie Royer

problem formulation

parent 29defe1f
\textbf{Problem Formulation:}
\item classes appear sequentially (or in batches) % $c_1,c_2,\dots,c_T$
\item each class $y$ has training images $X^y=\{x^y_1,\dots,x^y_{n_y}\}$
\item Pretrained model $M$ for task $T: X \rightarrow Y$
\item How to most efficiently fine-tune $M$ for new task $T': X' \rightarrow Y$ where $\mathcal{P}(X) \neq \mathcal{P}(X')$
\item \textbf{Baseline:} Fine-tuning
\item slow for large models
\item overfitting (small $X'$, many parameters)
\textbf{Example (PACS~\blue{[1]} dataset and variants)}
\centering \includegraphics[height=\pacsheight]{pacs_1.png}
\centering \scriptsize Photo
\centering \scriptsize \blue{Photo (base)}
\centering \includegraphics[height=\pacsheight]{pacs_2.png}
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment