Deming Zhai

Variation-Bounded Losses for Learning with Noisy Labels

Communication-Efficient Federated Learning via Model Update Distillation

Zero-Mean Regularized Spectral Contrastive Learning: Implicitly Mitigating Wrong Connections in Positive-Pair Graphs

$\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise

GM-DDPM: Denoising diffusion probabilistic models with Gaussian Mixture Noise