Controlling Neural Networks via Energy Dissipation

Moeller, Michael; Moellenhoff, Thomas; Cremers, Daniel

doi:10.1109/iccv.2019.00335

Cited by 11 publications

(8 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results by cVAE, LGD, and GM3 are comparable, at least visually. The ability to reconstruct tumours further indicates that cVAE does not miss out on important features that are not present in the training data, indicating a certain degree of robustness of the cVAE framework, so long as the signal is sufficiently strong, since many deep learning-based methods tend to miss the tumour due to the bias induced by tumour-free training data [57].…”

Section: Numerical Experiments and Discussionmentioning

confidence: 99%

Conditional Variational Autoencoder for Learned Image Reconstruction

2021

View full text Add to dashboard Cite

Learned image reconstruction techniques using deep neural networks have recently gained popularity and have delivered promising empirical results. However, most approaches focus on one single recovery for each observation, and thus neglect information uncertainty. In this work, we develop a novel computational framework that approximates the posterior distribution of the unknown image at each query observation. The proposed framework is very flexible: it handles implicit noise models and priors, it incorporates the data formation process (i.e., the forward operator), and the learned reconstructive properties are transferable between different datasets. Once the network is trained using the conditional variational autoencoder loss, it provides a computationally efficient sampler for the approximate posterior distribution via feed-forward propagation, and the summarizing statistics of the generated samples are used for both point-estimation and uncertainty quantification. We illustrate the proposed framework with extensive numerical experiments on positron emission tomography (with both moderate and low-count levels) showing that the framework generates high-quality samples when compared with state-of-the-art methods.

show abstract

Section: Numerical Experiments and Discussionmentioning

confidence: 99%

Conditional Variational Autoencoder for Learned Image Reconstruction

2021

View full text Add to dashboard Cite

show abstract

“…[13], or combining optimizers with networks that have been trained individually [14,15]. The recent work of Moeller et al [16] trains a network to predict descent directions to a given energy in order to give provable convergence results on the learned optimizer.…”

Section: Related Workmentioning

confidence: 99%

Training Auto-Encoder-Based Optimizers for Terahertz Image Reconstruction

Wong

Kahl

Bolívar

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Terahertz (THz) sensing is a promising imaging technology for a wide variety of different applications. Extracting the interpretable and physically meaningful parameters for such applications, however, requires solving an inverse problem in which a model function determined by these parameters needs to be fitted to the measured data. Since the underlying optimization problem is nonconvex and very costly to solve, we propose learning the prediction of suitable parameters from the measured data directly. More precisely, we develop a model-based autoencoder in which the encoder network predicts suitable parameters and the decoder is fixed to a physically meaningful model function, such that we can train the encoding network in an unsupervised way. We illustrate numerically that the resulting network is more than 140 times faster than classical optimization techniques while making predictions with only slightly higher objective values. Using such predictions as starting points of local optimization techniques allows us to converge to better local minima about twice as fast as optimization without the network-based initialization.

show abstract

“…Previous works have addressed similar problems by training networks that are safeguarded by a suitable cost function in order to guarantee bounds such as (3), see [1,2]. Unfortunately, the approach of Moeller et al [1] is limited to the case where d(Au, f ) is continuously differentiable in u, e.g. the investigated classical case of Gaussian noise where d(Au, f ) = ∥Au − f ∥ 2 .…”

Section: Introductionmentioning

confidence: 99%

Non-Smooth Energy Dissipating Networks

Dröge

Möllenhoff

Möller

2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Over the past decade, deep neural networks have been shown to perform extremely well on a variety of image reconstruction tasks. Such networks do, however, fail to provide guarantees about these predictions, making them difficult to use in safety-critical applications. Recent works addressed this problem by combining modeland learning-based approaches, e.g., by forcing networks to iteratively minimize a model-based cost function via the prediction of suitable descent directions. While previous approaches were limited to continuously differentiable cost functions, this paper discusses a way to remove the restriction of differentiability. We propose to use the Moreau-Yosida regularization of such costs to make the framework of energy dissipating networks applicable. We demonstrate our framework on two exemplary applications, i.e., safeguarding energy dissipating denoising networks to the expected distribution of the noise as well as enforcing binary constraints on bar-code deblurring networks to improve their respective performances.

show abstract

Controlling Neural Networks via Energy Dissipation

Cited by 11 publications

References 43 publications

Conditional Variational Autoencoder for Learned Image Reconstruction

Conditional Variational Autoencoder for Learned Image Reconstruction

Training Auto-Encoder-Based Optimizers for Terahertz Image Reconstruction

Non-Smooth Energy Dissipating Networks

Contact Info

Product

Resources

About