\begin{frame}
\frametitle{Topic 4: Shapley Values}

%\Large \\
	\LARGE \textbf{Chapter:} 9.2,9.5 and 9.6 + 2/3 papers \\
	\textbf{Supervisor:} Chiara Balestra (chiara.balestra@cs.uni-dortmund.de)\\

    %\begin{center}
        %\includegraphics[height=3cm]{illustrations/chiara.png}
    %\end{center}

    \begin{columns}
        %\begin{column}{.3\textwidth}
        %    \begin{center}
        %        \includegraphics[height=3cm]{illustrations/chiara.png}
        %    \end{center}
        %\end{column}
	\begin{column}{0.475\textwidth}
		Focus either on Computer Science
		\begin{itemize}
                        \item \small\textbf{Shapley Values for Feature Selection: The Good, the Bad, and the Axioms} (Fryer et al. 2020)
                        \item \textbf{Explaining Models by Propagating Shapley Values of Local Components} (Chen et al. 2020)
                        \item \textbf{GraphSVX: Shapley Value Explanations for Graph Neural Networks} (Duval et al. 2021)
	        \end{itemize}
	\end{column}
	\begin{column}{0.475\textwidth}
		Or on Medical Application
		\begin{itemize}
                        \item \small\textbf{Identifying mortality factors from Machine Learning using Shapley values – a case of COVID19} (Smith et al. 2021)
                        \item \textbf{Explaining multivariate molecular diagnostic tests via Shapley values} (Roder et al. 2021)
	        \end{itemize}
	\end{column}

    \end{columns}



\end{frame}