1 code implementation • 13 Aug 2024 • Jonas Linkerhägner, Cheng Shi, Ivan Dokmanić

In graph learning the graph and the node features both contain noisy information about the node labels.

1 code implementation • 28 Jul 2024 • Cheng Shi, Liming Pan, Ivan Dokmanić

A central question in deep learning is how deep neural networks (DNNs) learn features.

1 code implementation • 4 Mar 2024 • Valentin Debarnot, Vinith Kishore, Ricardo D. Righetto, Ivan Dokmanić

We introduce ICE-TIDE, a method for cryogenic electron tomography (cryo-ET) that simultaneously aligns observations and reconstructs a high-resolution volume.

1 code implementation • 1 Jan 2024 • AmirEhsan Khorashadizadeh, Valentin Debarnot, Tianlin Liu, Ivan Dokmanić

Deep learning is the current de facto state of the art in tomographic imaging.

no code implementations • 25 Jul 2023 • Vinith Kishore, Valentin Debarnot, Ivan Dokmanić

Cryo-electron tomography (cryoET) is a technique that captures images of biological samples at different tilts, preserving their native state as much as possible.

1 code implementation • 14 Jul 2023 • Cheng Shi, Maarten V. de Hoop, Ivan Dokmanić

Existing techniques relying on coarsely approximated, fixed wave speed models fail in this unexplored dense regime where the complexity of unknown wave speed cannot be ignored.

1 code implementation • 9 Jun 2023 • Liming Pan, Cheng Shi, Ivan Dokmanić

In this work, we propose a \textit{graph dynamics prior} (GDP) for relational inference.

no code implementations • 24 Apr 2023 • Anastasis Kratsios, Chong Liu, Matti Lassas, Maarten V. de Hoop, Ivan Dokmanić

Motivated by the developing mathematics of deep learning, we build universal functions approximators of continuous maps between arbitrary Polish metric spaces $\mathcal{X}$ and $\mathcal{Y}$ using elementary functions between Euclidean spaces as building blocks.

1 code implementation • 27 Feb 2023 • Antoine Maillard, Afonso S. Bandeira, David Belius, Ivan Dokmanić, Shuta Nakajima

Recent work connects this problem to spherical integral geometry giving rise to a conjectured sharp injectivity threshold for $\alpha = \frac{m}{n}$ by studying the expected Euler characteristic of a certain random set.

1 code implementation • 8 Jan 2023 • AmirEhsan Khorashadizadeh, Vahid Khorashadizadeh, Sepehr Eskandari, Guy A. E. Vandenbosch, Ivan Dokmanić

Unlike supervised methods that necessitate both scattered fields and target permittivities, our method only requires the target permittivities for training; it can then be used with any experimental setup.

1 code implementation • 26 Dec 2022 • Cheng Shi, Liming Pan, Hong Hu, Ivan Dokmanić

Motivated by experimental observations of ``transductive'' double descent in key networks and datasets, we use analytical tools from statistical physics and random matrix theory to precisely characterize generalization in simple graph convolution networks on the contextual stochastic block model.

1 code implementation • 20 Dec 2022 • AmirEhsan Khorashadizadeh, Anadi Chaman, Valentin Debarnot, Ivan Dokmanić

Our answer is FunkNN -- a new convolutional network which learns how to reconstruct continuous images at arbitrary coordinates and can be applied to any image dataset.

1 code implementation • 8 Dec 2022 • AmirEhsan Khorashadizadeh, Ali Aghababaei, Tin Vlašić, Hieu Nguyen, Ivan Dokmanić

Inverse medium scattering solvers generally reconstruct a single solution without an associated measure of uncertainty.

1 code implementation • 18 Nov 2022 • Sidharth Gupta, Konik Kothari, Valentin Debarnot, Ivan Dokmanić

We propose a differentiable imaging framework to address uncertainty in measurement coordinates such as sensor locations and projection angles.

1 code implementation • NeurIPS 2023 • Anastasis Kratsios, Valentin Debarnot, Ivan Dokmanić

We derive embedding guarantees for feature maps implemented by small neural networks called \emph{probabilistic transformers}.

no code implementations • 12 Sep 2022 • Valentin Debarnot, Vinith Kishore, Cheng Shi, Ivan Dokmanić

We illustrate our graph denoising framework on regular synthetic graphs and then apply it to single-particle cryo-EM where the measurements are corrupted by very high levels of noise.

1 code implementation • 6 Jul 2022 • Shuai Huang, Mona Zehni, Ivan Dokmanić, Zhizhen Zhao

Unknown-view tomography (UVT) reconstructs a 3D density map from its 2D projections at unknown, random orientations.

no code implementations • 4 Jun 2022 • Tin Vlašić, Hieu Nguyen, AmirEhsan Khorashadizadeh, Ivan Dokmanić

In this paper, we introduce an implicit neural representation-based framework for solving the inverse obstacle scattering problem in a mesh-free fashion.

1 code implementation • 15 Apr 2022 • AmirEhsan Khorashadizadeh, Konik Kothari, Leonardo Salsi, Ali Aghababaei Harandi, Maarten de Hoop, Ivan Dokmanić

Most deep learning models for computational imaging regress a single reconstructed image.

1 code implementation • ICLR 2022 • Liming Pan, Cheng Shi, Ivan Dokmanić

Instead of extracting transition probabilities from the original graph, it computes the transition matrix of a "predictive" latent graph by applying attention to learned features; this may be interpreted as feature-sensitive topology fingerprinting.

Ranked #1 on Link Prediction on Pubmed

no code implementations • 8 Oct 2021 • Michael Puthawala, Matti Lassas, Ivan Dokmanić, Maarten de Hoop

We show that in general, injective flows between $\mathbb{R}^n$ and $\mathbb{R}^m$ universally approximate measures supported on images of extendable embeddings, which are a subset of standard embeddings: when the embedding dimension m is small, topological obstructions may preclude certain manifolds as admissible targets.

no code implementations • ICLR 2022 • Anastasis Kratsios, Behnoosh Zamanlooy, Tianlin Liu, Ivan Dokmanić

Many practical problems need the output of a machine learning model to satisfy a set of constraints, $K$.

1 code implementation • 9 May 2021 • Anadi Chaman, Ivan Dokmanić

Convolutional neural networks lack shift equivariance due to the presence of downsampling layers.

1 code implementation • 20 Feb 2021 • Konik Kothari, AmirEhsan Khorashadizadeh, Maarten de Hoop, Ivan Dokmanić

We propose injective generative models called Trumpets that generalize invertible normalizing flows.

1 code implementation • 6 Feb 2021 • Dalia El Badawy, Viktor Larsson, Marc Pollefeys, Ivan Dokmanić

We look at the general case where neither the emission times of the sources nor the reference time frames of the receivers are known.

1 code implementation • 1 Feb 2021 • Sidharth Gupta, Ivan Dokmanić

We address the phase retrieval problem with errors in the sensing vectors.

2 code implementations • CVPR 2021 • Anadi Chaman, Ivan Dokmanić

Thanks to the use of convolution and pooling layers, convolutional neural networks were for a long time thought to be shift-invariant.

2 code implementations • 25 Nov 2020 • Tianlin Liu, Anadi Chaman, David Belius, Ivan Dokmanić

To close the performance gap, we thus propose a multiscale convolutional dictionary structure.

no code implementations • 17 Jun 2020 • Puoya Tabaghi, Jianhao Peng, Olgica Milenkovic, Ivan Dokmanić

To study this question, we introduce the notions of the \textit{ordinal capacity} of a target space form and \emph{ordinal spread} of the similarity measurements.

no code implementations • 15 Jun 2020 • Michael Puthawala, Konik Kothari, Matti Lassas, Ivan Dokmanić, Maarten de Hoop

Injectivity plays an important role in generative models where it enables inference; in inverse problems and compressed sensing with generative priors it is a precursor to well posedness.

1 code implementation • NeurIPS 2020 • Konik Kothari, Maarten de Hoop, Ivan Dokmanić

We propose a general physics-based deep learning architecture for wave-based imaging problems.

no code implementations • 18 May 2020 • Puoya Tabaghi, Ivan Dokmanić

Hyperbolic space is a natural setting for mining and visualizing data with hierarchical structure.

1 code implementation • 4 Nov 2019 • Sidharth Gupta, Rémi Gribonval, Laurent Daudet, Ivan Dokmanić

Our method simplifies the calibration of optical transmission matrices from a quadratic to a linear inverse problem by first recovering the phase of the measurements.

1 code implementation • 9 Oct 2019 • Benjamín Béjar, Ivan Dokmanić, René Vidal

In this paper we study the proximal operator of the mixed $\ell_{1,\infty}$ matrix norm and show that it can be computed in closed form by applying the well-known soft-thresholding operator to each column of the matrix.

1 code implementation • NeurIPS 2019 • Sidharth Gupta, Rémi Gribonval, Laurent Daudet, Ivan Dokmanić

A signal of interest $\mathbf{\xi} \in \mathbb{R}^N$ is mixed by a random scattering medium to compute the projection $\mathbf{y} = \mathbf{A} \mathbf{\xi}$, with $\mathbf{A} \in \mathbb{C}^{M \times N}$ being a realization of a standard complex Gaussian iid random matrix.

1 code implementation • 14 Feb 2019 • Shuai Huang, Sidharth Gupta, Ivan Dokmanić

We tackle the problem of recovering a complex signal $\boldsymbol x\in\mathbb{C}^n$ from quadratic measurements of the form $y_i=\boldsymbol x^*\boldsymbol A_i\boldsymbol x$, where $\boldsymbol A_i$ is a full-rank, complex random measurement matrix whose entries are generated from a rotation-invariant sub-Gaussian distribution.

Information Theory Information Theory

no code implementations • 29 Jan 2019 • Puoya Tabaghi, Maarten de Hoop, Ivan Dokmanić

We study the learnability of a class of compact operators known as Schatten--von Neumann operators.

1 code implementation • 25 Nov 2018 • Mona Zehni, Shuai Huang, Ivan Dokmanić, Zhizhen Zhao

For a point source model, we show that these features reveal geometric information about the model such as the radial and pairwise distances.

1 code implementation • ICLR 2019 • Sidharth Gupta, Konik Kothari, Maarten V. de Hoop, Ivan Dokmanić

We show that in this case the common approach to directly learn the mapping from the measured data to the reconstruction becomes unstable.

1 code implementation • 6 Apr 2018 • Shuai Huang, Ivan Dokmanić

Our method is the first practical approach to solve the large-scale noisy beltway problem where the points lie on a loop.

2 code implementations • 11 Oct 2017 • Robin Scheibler, Eric Bezzam, Ivan Dokmanić

We present pyroomacoustics, a software package aimed at the rapid development and testing of audio array processing algorithms.

Sound Audio and Speech Processing

no code implementations • 18 Sep 2016 • Ivan Dokmanić, Joan Bruna, Stéphane Mallat, Maarten de Hoop

We propose a new approach to linear ill-posed inverse problems.

Computational Engineering, Finance, and Science

Cannot find the paper you are looking for? You can
Submit a new open access paper.

Contact us on:
hello@paperswithcode.com
.
Papers With Code is a free resource with all data licensed under CC-BY-SA.