Temperature scaling has been widely used as an effective approach to con...
Deep image classifiers often perform poorly when training data are heavi...
Analog computing hardware has gradually received more attention by the
r...
Label hierarchies widely exist in many vision-related problems, ranging ...
Model robustness has been an important issue, since adding small adversa...
Learning with a primary objective, such as softmax cross entropy for
cla...
Recent breakthroughs in Neural Architectural Search (NAS) have achieved
...
Recent studies on neural architecture search have shown that automatical...