Unsupervised Neural Universal Denoiser for Finite-Input General-Output Noisy Channel
This addresses denoising in practical scenarios with known channel densities, offering improved efficiency and performance over prior methods, though it is incremental in its approach.
The paper tackles the problem of universal denoising for finite-input, general-output channels by proposing Gen-CUDE, a neural network-based method that matches the performance of the best sliding window denoiser, achieving significantly better results in experiments with synthetic and real data.
We devise a novel neural network-based universal denoiser for the finite-input, general-output (FIGO) channel. Based on the assumption of known noisy channel densities, which is realistic in many practical scenarios, we train the network such that it can denoise as well as the best sliding window denoiser for any given underlying clean source data. Our algorithm, dubbed as Generalized CUDE (Gen-CUDE), enjoys several desirable properties; it can be trained in an unsupervised manner (solely based on the noisy observation data), has much smaller computational complexity compared to the previously developed universal denoiser for the same setting, and has much tighter upper bound on the denoising performance, which is obtained by a theoretical analysis. In our experiments, we show such tighter upper bound is also realized in practice by showing that Gen-CUDE achieves much better denoising results compared to other strong baselines for both synthetic and real underlying clean sequences.