This work proposes a Momentum-Enabled Kronecker-Factor-Based Optimizer Using Rank-1 Updates, called MKOR, that improves the training time and convergence properties of deep neural networks (DNNs). Second-order techniques, while enjoying higher convergence rates vs first-order counterparts, have cubic complexity with respect to either the model size and/or the training batch size.

Unsupervised video domain adaptation is a practical yet challenging task.

L for LM; D for diffusion), an LM-guided diffusion model that generates music audios of state-of-the-art quality meanwhile reducing 95.7% to 99.6% forward