Neural Universal Discrete Denoiser
This addresses the challenge of denoising discrete data in practical settings where clean labels are unavailable, offering an incremental improvement over existing methods.
The paper tackles the problem of discrete denoising without requiring clean training data by introducing pseudo-labels and a novel objective function, resulting in Neural DUDE, which significantly outperforms previous state-of-the-art methods in several applications with a systematic hyperparameter selection rule.
We present a new framework of applying deep neural networks (DNN) to devise a universal discrete denoiser. Unlike other approaches that utilize supervised learning for denoising, we do not require any additional training data. In such setting, while the ground-truth label, i.e., the clean data, is not available, we devise "pseudo-labels" and a novel objective function such that DNN can be trained in a same way as supervised learning to become a discrete denoiser. We experimentally show that our resulting algorithm, dubbed as Neural DUDE, significantly outperforms the previous state-of-the-art in several applications with a systematic rule of choosing the hyperparameter, which is an attractive feature in practice.