1 code implementation • 4 Oct 2023 • Zahra Kadkhodaie, Florentin Guth, Eero P. Simoncelli, Stéphane Mallat
Finally, we show that when trained on regular image classes for which the optimal basis is known to be geometry-adaptive and harmonic, the denoising performance of the networks is near-optimal.
1 code implementation • 6 Mar 2023 • Zahra Kadkhodaie, Florentin Guth, Stéphane Mallat, Eero P Simoncelli
We instantiate this model using convolutional neural networks (CNNs) with local receptive fields, which enforce both the stationarity and Markov properties.
no code implementations • NeurIPS 2021 • Zahra Kadkhodaie, Eero Simoncelli
Two recent lines of work – Denoising Score Matching and Plug-and-Play – propose methodologies for drawing samples from this implicit prior and using it to solve inverse problems, respectively.
1 code implementation • 27 Jul 2020 • Zahra Kadkhodaie, Eero P. Simoncelli
Here, we develop a robust and general methodology for making use of this implicit prior.
no code implementations • NeurIPS Workshop Deep_Invers 2019 • Zahra Kadkhodaie, Sreyas Mohan, Eero P. Simoncelli, Carlos Fernandez-Granda
Here, however, we show that bias terms used in most CNNs (additive constants, including those used for batch normalization) interfere with the interpretability of these networks, do not help performance, and in fact prevent generalization of performance to noise levels not including in the training data.
1 code implementation • ICLR 2020 • Sreyas Mohan, Zahra Kadkhodaie, Eero P. Simoncelli, Carlos Fernandez-Granda
In contrast, a bias-free architecture -- obtained by removing the constant terms in every layer of the network, including those used for batch normalization-- generalizes robustly across noise levels, while preserving state-of-the-art performance within the training range.