Unlike projection based algorithms [14,32] though, momentum does not perform well with FW.

We show results on CIFAR-100 LT with an imbalance factor (β)of100and200.

Several techniques based on re-weighting and margin adjustment of loss are often used toenhance theperformance ofneural networks, particularly onminority classes.

The strong correlation between neurons or filters can significantly weaken the generalization ability of neural networks.

InInternational Journalof Parallel, Emergentand Distributed Systems, volume 27, pages 387-408, 2012.