Open TangJiakai opened 1 year ago
Should we get the opposite conclusion? i.e., MPt >= MP{t+1}
Another question: Clean samples in this paper seem to only consider positive sample interactions (the calculation of m_t), without considering whether negative sample interactions are clean?
Should we get the opposite conclusion? i.e., MPt >= MP{t+1}