English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We propose a novel iterative channel estimation (ICE) algorithm thatessentially removes the critical known noisy channel assumption for universaldiscrete denoising problem. Our algorithm is based on Neural DUDE (N-DUDE), arecently proposed neural network-based discrete denoiser, and it estimates thechannel transition matrix as well as the neural network parameters in analternating manner until convergence. While we do not make any probabilisticassumption on the underlying clean data, our ICE resemblesExpectation-Maximization (EM) with variational approximation, and it takesadvantage of the property of N-DUDE being locally robust around the truechannel. With extensive experiments on several radically different types ofdata, we show that the ICE equipped N-DUDE (dubbed as ICE-N-DUDE) can perform\emph{universally} well regardless of the uncertainties in both the channel andthe clean source. Moreover, we show ICE-N-DUDE becomes extremely robust to itshyperparameters and significantly outperforms the strong baseline that can dealwith the channel uncertainties for denoising, the widely used Baum-Welch (BW)algorithm for hidden Markov models (HMM).

Iterative Channel Estimation for Discrete Denoising under Channel Uncertainty