Binarized Weight Networks for Inverse Problems

Date

2020-01-01

Author

Ozkan, Savas
Becek, Kadircan
Inci, Alperen
Kutukcu, Basar
Ugurcali, Faruk
Kaya, Mete Can
Akar, Gözde

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

146
views

0
downloads

In this paper, we present a binarized neural network structure for inverse problems. In this structure, memory requirements and computation time are significantly reduced with a negligible performance drop compared to full-precision models. For this purpose, a unique architecture is proposed based on a residual learning. Precisely, it opts to reconstruct only the error between input and output images, which is eventually centralized the responses around zero. To this end, this provides several advantages for binary representation and manifold space is adopted to learn with binarized networks. Experiments are conducted on three different inverse problems as super-resolution, denoising and deblurring problems for various datasets. The results validate the success of the method.

Subject Keywords

Inverse problems, Super-Resolution, Denoising, Deblurring, Binary Networks

URI

https://hdl.handle.net/11511/96529

DOI

https://doi.org/10.1109/siu49456.2020.9302129

Conference Name

28th Signal Processing and Communications Applications Conference (SIU)

Collections

Department of Electrical and Electronics Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Domain-Structured Chaos in a Hopfield Neural Network Akhmet, Marat (World Scientific Pub Co Pte Lt, 2019-12-30) In this paper, we provide a new method for constructing chaotic Hopfield neural networks. Our approach is based on structuring the domain to form a special set through the discrete evolution of the network state variables. In the chaotic regime, the formed set is invariant under the system governing the dynamics of the neural network. The approach can be viewed as an extension of the unimodality technique for one-dimensional map, thereby generating chaos from higher-dimensional systems. We show that the dis...
Fully computable convergence analysis of discontinous Galerkin finite element approximation with an arbitrary number of levels of hanging nodes Özışık, Sevtap; Kaya Merdan, Songül; Riviere, Beatrice M.; Department of Mathematics (2012) In this thesis, we analyze an adaptive discontinuous finite element method for symmetric second order linear elliptic operators. Moreover, we obtain a fully computable convergence analysis on the broken energy seminorm in first order symmetric interior penalty discontin- uous Galerkin finite element approximations of this problem. The method is formulated on nonconforming meshes made of triangular elements with first order polynomial in two di- mension. We use an estimator which is completely free of unknow...
Inverse Sturm-Liouville Systems over the whole Real Line Altundağ, Hüseyin; Taşeli, Hasan; Department of Mathematics (2010) In this thesis we present a numerical algorithm to solve the singular Inverse Sturm-Liouville problems with symmetric potential functions. The singularity, which comes from the unbounded domain of the problem, is treated by considering the limiting case of the associated problem on the symmetric finite interval. In contrast to regular problems which are considered on a finite interval the singular inverse problem has an ill-conditioned structure despite of the limiting treatment. We use the regularization t...
Discrete linear Hamiltonian systems: Lyapunov type inequalities, stability and disconjugacy criteria Zafer, Ağacık (2012-12-15) In this paper, we first establish new Lyapunov type inequalities for discrete planar linear Hamiltonian systems. Next, by making use of the inequalities, we derive stability and disconjugacy criteria. Stability criteria are obtained with the help of the Floquet theory, so the system is assumed to be periodic in that case.
Stability analysis of recurrent neural networks with piecewise constant argument of generalized type Akhmet, Marat; Yılmaz, Elanur (2010-09-01) In this paper, we apply the method of Lyapunov functions for differential equations with piecewise constant argument of generalized type to a model of recurrent neural networks (RNNs). The model involves both advanced and delayed arguments. Sufficient conditions are obtained for global exponential stability of the equilibrium point. Examples with numerical simulations are presented to illustrate the results.

Citation Formats

S. Ozkan et al., “Binarized Weight Networks for Inverse Problems,” presented at the 28th Signal Processing and Communications Applications Conference (SIU), ELECTR NETWORK, 2020, Accessed: 00, 2022. [Online]. Available: https://hdl.handle.net/11511/96529.