Table 1. Important Definitions.

Symbols	Definitions
SD	The source domain dataset.
TD	The target domain dataset.
Xi	i-th bags, which represent a protein.
xj	-th instance in a bag.
	j-th instance in bag Xi.
	The average of all the instances in bag Xi.
Yi	Represent the Gene Ontology terms assigned to Xi.
	k-th element in Yi.
ci	The center of Xi.
D(xi, ci)	The square of the Mahalanobis distance between instance xi and ci.
D(Xi, Xj)	The square of the Mahalanobis distance between bags Xi and Xj.
A	The learned Mahalanobis distance metric.
	The loss corresponding to traditional Multi-Instance Metric Learning.
	The expected loss.
δS	A constant to limit the minimum distance between the center of the bag and the instance in the bag.
δD	A constant to limit the maximum distance between bags from different class.
ξ, ζ	Two slack vectors to improve the robustness of the algorithm.
ω	The weight vector of bags.