Updated REEzultz

content/experiments/experiments.tex

@@ -150,7 +150,7 @@ Yeast3       & \textbf{0.879{\tiny{$\pm$0.032}}} & 0.861{\tiny{$\pm$0.070}} & 0.
 
 \subsection{CIFAR-10N: Label-Noise Robustness}\label{sec:results:cifar10n}
 
-\Cref{tab:results:cifar10n} shows AUC under three noise levels on CIFAR-10N. Under \emph{clean} labels all methods reach AUC $\geq 0.972$, with \textsc{ROLL}-AOC at the top (0.980). As noise increases the advantage of \textsc{ROLL}-AOC grows: under \emph{aggre} noise it leads by 0.002 over the next-best method (GCE, 0.932) while other methods degrade more significantly, and under \emph{worse} noise, the margin widens to 0.012 over MAE (0.809) and 0.083 over BCE-W (0.738). Observing degradation relative to the clean scenario, \textsc{ROLL}-AOC degraded the least ($-0.159$ relative to clean results), with MAE second ($-0.167$).
+\Cref{tab:results:cifar10n} shows AUC under three noise levels on CIFAR-10N. Under \emph{clean} labels all methods reach AUC $\geq 0.973$, with \textsc{ROLL}-AOC at the top (0.980). As noise increases the advantage of \textsc{ROLL}-AOC grows: under \emph{aggre} noise it leads by 0.001 over the next-best method (LibAUC, 0.931) and 0.002 over GCE (0.930), and under \emph{worse} noise the margin widens to 0.008 over GCE (0.811) and 0.013 over BCE-W (0.806). Observing degradation relative to the clean scenario, \textsc{ROLL}-AOC degraded the least ($-0.161$ relative to clean results), while the remaining methods degraded between $-0.163$ (Asym.) and $-0.169$ (BCE-W and LibAUC).
 
 \begin{table}[htbp]
 \centering
@@ -161,9 +161,9 @@ Yeast3       & \textbf{0.879{\tiny{$\pm$0.032}}} & 0.861{\tiny{$\pm$0.070}} & 0.
 \toprule
 Noise & \textsc{ROLL}-AOC & BCE-W & Focal & Asym. & GCE & LibAUC & MAE \\
 \midrule
-Clean & \textbf{0.980{\tiny{$\pm$0.001}}} & 0.977{\tiny{$\pm$0.002}} & 0.977{\tiny{$\pm$0.000}} & 0.972{\tiny{$\pm$0.003}} & 0.976{\tiny{$\pm$0.002}} & 0.979{\tiny{$\pm$0.000}} & 0.976{\tiny{$\pm$0.002}} \\
-Aggre. & \textbf{0.934{\tiny{$\pm$0.002}}} & 0.924{\tiny{$\pm$0.003}} & 0.912{\tiny{$\pm$0.014}} & 0.910{\tiny{$\pm$0.004}} & 0.932{\tiny{$\pm$0.005}} & 0.921{\tiny{$\pm$0.004}} & 0.930{\tiny{$\pm$0.002}} \\
-Worse & \textbf{0.821{\tiny{$\pm$0.001}}} & 0.738{\tiny{$\pm$0.009}} & 0.734{\tiny{$\pm$0.009}} & 0.720{\tiny{$\pm$0.004}} & 0.774{\tiny{$\pm$0.005}} & 0.752{\tiny{$\pm$0.006}} & 0.809{\tiny{$\pm$0.004}} \\
+Clean & \textbf{0.980{\tiny{$\pm$0.001}}} & 0.975{\tiny{$\pm$0.003}} & 0.973{\tiny{$\pm$0.002}} & 0.973{\tiny{$\pm$0.003}} & 0.976{\tiny{$\pm$0.002}} & 0.979{\tiny{$\pm$0.002}} & 0.974{\tiny{$\pm$0.003}} \\
+Aggre. & \textbf{0.932{\tiny{$\pm$0.002}}} & 0.927{\tiny{$\pm$0.002}} & 0.922{\tiny{$\pm$0.003}} & 0.927{\tiny{$\pm$0.002}} & 0.930{\tiny{$\pm$0.001}} & 0.931{\tiny{$\pm$0.002}} & 0.929{\tiny{$\pm$0.002}} \\
+Worse & \textbf{0.819{\tiny{$\pm$0.000}}} & 0.806{\tiny{$\pm$0.002}} & 0.809{\tiny{$\pm$0.004}} & 0.810{\tiny{$\pm$0.003}} & 0.811{\tiny{$\pm$0.001}} & 0.810{\tiny{$\pm$0.004}} & 0.810{\tiny{$\pm$0.001}} \\
 \bottomrule
 \end{tabular}%
 }