End-to-end speech separation with improved phase reconstruction.
problem Cocktail party problem: separating multiple speakers in a single-channel recording.
method End-to-end approach using deep learning with unfolded phase reconstruction iterations and novel activation functions.
result State-of-the-art performance on SI-SDR and SDR metrics.
New method uses generative models to improve phase retrieval stability.
problem Improving stability of solutions in phase retrieval problems.
method Unified reconstruction approach using generative models to mitigate overfitting.
result Mitigates overfitting to generative model for varying noise levels.
Generative adversarial network improves signal reconstruction from magnitude spectrograms.
problem Reconstructing a time-domain signal from a magnitude spectrogram.
method Deep neural network and generative adversarial network approach.
result Our method reconstructs signals faster with higher quality than the Griffin-Lim method.
Belief Propagation outperforms other algorithms in reconstructing binary symmetric channel trees.
problem Reconstructing binary symmetric channel trees with bounded memory.
method Combining recursive reconstruction, information theory, and optimal transport.
result Any recursive algorithm with bounded memory for the reconstruction problem on binary symmetric channel trees has a phase transition strictly below the Belief Propagation threshold.
New method reconstructs signals from modulo observations.
problem Reconstructing signals from under-determined modulo observations.
method Proposes a novel algorithm inspired by phase retrieval for under-determined signal reconstruction.
result Successfully recovers signals with improved performance over existing methods.
We propose a new algorithm to learn a dictionary for reconstructing and sparsely encoding signals from measurements without phase. Specifically, we consider the task of estimating a two-dimensional image from squared-magnitude measurements of a complex-valued linear transformation of the original image. Several recent …
Describes reconstructing Poisson structures from Lie group actions.
problem Reconstructing invariant Poisson structures from Lie group actions.
method Describes reconstruction of invariant Poisson structures from canonical actions of compact Lie groups on fibered phase spaces.
result Derives symmetry properties of Wong's type equations from main results.
Hard phase in inference problems is glassy and hard to reconstruct.
problem Hard phase in inference problems that are hard to solve algorithmically.
method Study of metastable states and their entropy in low-rank matrix factorization.
result AMP algorithm performance is not improved by considering glassy states.
Deep learning reconstructs ultra-short laser pulses.
problem Characterization of ultra-short laser pulses (femtosecond to attosecond).
method Deep neural network technique.
result First deep learning method to reconstruct ultra-short optical pulses.
We study phase retrieval from magnitude measurements of an unknown signal as an algebraic estimation problem. Indeed, phase retrieval from rank-one and more general linear measurements can be treated in an algebraic way. It is verified that a certain number of generic rank-one or generic linear measurements are suffici…
Quantization-aware phase retrieval algorithm improves signal reconstruction accuracy.
problem Reconstructing signals from quantized phase measurements.
method Developed a rank-1 projection algorithm with consistency criterion using one-sided quadratic cost.
result The algorithm achieves higher reconstruction accuracy and is closer to the Cramér-Rao lower bound.
Study reconstructs hidden perfect matchings in random graphs with specific edge weights.
problem Reconstructing hidden perfect matchings in random weighted bipartite graphs.
method Analyzes the maximum likelihood estimator for matching reconstruction under different probability distributions of edge weights.
result Sharp threshold and infinite-order phase transition in reconstruction error for different probability distributions.
Deep learning removes aliasing in MRI scans with fast computation.
problem High computational costs in MR scan reconstruction.
method Deep residual learning networks for magnitude and phase networks.
result Deep learning successfully removes aliasing artifacts from MRI scans.
DeepPhaseCut uses neural networks to improve Fourier phase retrieval.
problem Fourier phase retrieval from magnitude data.
method Unsupervised feed-forward neural network with cycleGAN training.
result Outperforms existing methods in Fourier phase retrieval.
Paper proposes RAN for better anomaly detection in time series data.
problem Anomaly detection algorithms often fail to accurately detect anomalies due to incomplete reconstruction of anomaly data.
method RAN uses adversarial learning and latent vector-constrained Autoencoder to ensure consistent reconstruction of anomaly data.
result RAN outperforms other algorithms in detecting meaningful anomalies with higher AUC-ROC scores.
A neural network learns phase space properties for time series analysis.
problem Lack of consistency and robustness in estimating embedding parameters.
method Forgetting mechanism neural network to learn phase space properties.
result Neural network approach is competitive or superior to state-of-the-art strategies.
Study reconstructs network interactions from oscillator dynamics data.
problem Reconstructing network interactions from observed oscillator dynamics.
method Machine learning methods applied to phase-oscillator networks.
result Reconstruction of network interactions and intrinsic dynamics parameters.
In this paper, we carry a detailed study of mechanical systems with configuration space Q⟶Q/G for which the base Q/G variables are being controlled. The overall system's motion is considered to be induced from the base one due to the presence of general non-holonomic constraints. It is shown that the…
Diffusion models reveal a phase transition in reconstructing high-level features.
problem Understanding the hierarchical structure of natural data.
method Study of hierarchical generative models of data using diffusion models.
result The backward diffusion process shows a phase transition at a threshold time, where high-level features suddenly drop in reconstructibility.
The paper tackles the problem of recovering network history from structure, identifying a phase transition in recoverability.
problem Recovering the history of complex networks from their current structure.
method Bayesian formulation, sequential Monte Carlo algorithm, heuristic approach.
result Identification of a phase transition in the quality of reconstructed history.
Paper analyzes latent space geometry in generative models using Fisher information.
problem Understanding the structure of latent spaces in generative models.
method Reconstructs Fisher information metric from generated samples and posterior distribution.
result Reveals fractal structure and abrupt changes in Fisher metric at phase boundaries.
New method improves sparse signal reconstruction using 1RSB-AMP.
problem Sparse signal reconstruction with improved accuracy.
method Developed 1RSB-AMP and 1RSB-SE for SCAD penalty minimization.
result 1RSB-AMP achieves improved reconstruction compared to RS-AMP.
Method generates dense fields from sparse measurements without needing spatial statistics or examples.
problem Generating dense physical fields from sparse measurements.
method Introduces a differentiable numerical simulator into neural network training.
result Superior results on fluid mechanics problems compared to statistical and neural network methods.
New layers estimate complex time-frequency masks without phase wrapping issues.
problem Lack of phase estimation in deep learning-based speech enhancement and source separation.
method Proposes magbook, phasebook, and combook layers for complex mask estimation.
result Match state-of-the-art performance on speaker separation datasets.
Neural net reconstructs dark matter density from halo velocities.
problem Reconstructing local dark matter density from halo velocities.
method Hybrid architecture combining U-Net and DeepSets.
result Hybrid network recovers density amplitudes and phases better than U-Net.
Mask-reconstruction pretraining helps in downstream tasks by capturing more semantic features.
problem How mask-reconstruction pretraining helps in downstream tasks and why it surpasses supervised learning.
method Theoretical analysis and experimental validation of mask-reconstruction pretraining (MRP) on auto-encoders.
result MRP provably captures more semantic features than supervised learning, leading to better performance in downstream tasks.
SAPAG attacks distributed learning by reconstructing true training data from gradients.
problem Privacy attacks on distributed learning systems through gradients.
method SAPAG uses a Gaussian kernel-based gradient difference distance measure.
result SAPAG can reconstruct training data on various DNNs and at different training phases.
Gradient descent with small random init mimics spectral methods for low-rank matrix recovery.
problem Reconstructing a low-rank matrix from few measurements.
method Gradient descent with small random initialization followed by a few iterations.
result Gradient descent from small random init converges to a well-generalizing solution.
In this paper we study the property of phase retrievability by redundant sysems of vectors under perturbations of the frame set. Specifically we show that if a set $\fc$ of m vectors in the complex Hilbert space of dimension n allows for vector reconstruction from magnitudes of its coefficients, then there is a pertu…
Hadamard Wirtinger Flow recovers sparse signals from fewer measurements.
problem Reconstructing sparse signals from magnitude-only measurements.
method Gradient descent with Hadamard parametrization (HWF).
result A single step of HWF recovers support from k(xmax∗)−2 samples. A new CNN-based algorithm improves Fourier ptychography for faster, more robust image reconstruction.
problem Slow and inefficient Fourier ptychography reconstruction under system aberrations.
method A CNN-based iterative phase retrieval algorithm trained on GPUs.
result Significantly faster and more robust image reconstruction under system aberrations.
The paper extends a neural model to estimate uncertainty in multi-interaction trajectory reconstruction.
problem Lack of uncertainty estimation in neural models for multi-interaction trajectory reconstruction.
method Extended Factorised Neural Relational Inference model to output mean and standard deviation for each component of the phase space vector, using various loss functions.
result Demonstrated the importance of physical meaning of variables and existence of local minima during training.
Deep model generates high-quality speech from spectrograms.
problem Speech reconstruction from spectrograms.
method Deep generative model with Gaussian and von Mises distributions for magnitude and phase, variational autoencoder framework.
result Generated speech has high perceptual quality and intelligibility.
We employ unsupervised machine learning techniques to learn latent parameters which best describe states of the two-dimensional Ising model and the three-dimensional XY model. These methods range from principal component analysis to artificial neural network based variational autoencoders. The states are sampled using …
This paper resolves the all-or-nothing phase transition in graph matching.
problem Recovering vertex correspondence between edge-correlated random graphs.
method Analysis of mutual information, truncated second-moment computation, and maximum likelihood estimator.
result Sharp thresholds for correct matching in both dense and sparse graphs.
Paper tackles sparse phase retrieval with a novel Bayesian approach.
problem Sparse phase retrieval from magnitude-only data.
method Quasi-Bayesian approach using a scaled Student distribution.
result Achieves minimax-optimal convergence rates under sub-exponential noise.
Study reconstructs Riemannian metric from Cherenkov radiation in complex media.
problem Reconstructing internal geometry of inhomogeneous anisotropic targets.
method Mathematical model of waves in medium, including vector-valued wave operator and phase velocity.
result Riemannian metric inside a bounded region can be reconstructed from boundary measurements of Cherenkov radiation.
CycleQSM uses deep learning to accurately map tissue magnetic susceptibility without needing paired data.
problem Accurately mapping magnetic susceptibility values from phase images using QSM.
method Unsupervised deep learning approach using physics-informed cycleGAN.
result The method provides more accurate QSM maps compared to existing deep learning approaches.
New framework UIFV reconstructs private features in VFL without model details.
problem Privacy risks in VFL where adversaries can reconstruct sensitive features.
method Unified InverNet Framework (UIFV) that uses intermediate feature data.
result Significantly outperforms state-of-the-art techniques in attack precision.
Machine learning reconstructs aerodynamic forces from noisy data.
problem Accurately modeling aerodynamic forces with limited or noisy data.
method Physics-informed Gaussian processes trained on noisy structural responses.
result Strong agreement between true and predicted aerodynamic loads.
Paper studies non-tight reconstruction threshold in a 4-state model with different in/out block mutations.
problem Non-tight reconstruction threshold in a 4-state symmetric model with different in-block and out-block mutations.
method Inspired by the q1+q2 stochastic block model, rigorously analyzes conditions for non-tightness of the reconstruction threshold. result Rigorously gives conditions for the non-tightness of the reconstruction threshold in a 4-state symmetric model.
Autoencoders misidentify anomalies due to data topology.
problem Autoencoders fail to accurately identify anomalies in data with nontrivial topology.
method Illustrative low-dimensional examples and analysis of autoencoder behavior in latent space.
result Topology of the dataset affects autoencoder performance, leading to misidentification of anomalies.
We present a generalization of Minkowski's classic theorem on the reconstruction of tetrahedra from algebraic data to homogeneously curved spaces. Euclidean notions such as the normal vector to a face are replaced by Levi-Civita holonomies around each of the tetrahedron's faces. This allows the reconstruction of both s…
Proposes a new method for MR image reconstruction using unsupervised deep learning.
problem Compensating for missing k-space data in MR images.
method Learns the probability distribution of fully sampled MR images using VAE and uses it as an explicit prior term in reconstruction.
result Produces high-quality reconstructions with low RMSE values, outperforming other methods.
Graph networks improve particle reconstruction in irregular detectors.
problem Handling irregular particle-detector geometries in particle reconstruction.
method Introduce distance-weighted graph network architectures (GarNet, GravNet layers) for irregular geometry detectors.
result The proposed graph networks provide equally performing or less resource-demanding solutions compared to existing methods.
Paper uses TDA to assess cryptocurrency risk by measuring phase space instability.
problem Traditional risk measures fail to capture market dynamics' geometric structure.
method Applied Takens' Delay Embedding Theorem to generate point cloud, computed persistent homology groups, defined Topological Persistence Norm.
result Proposed leverage calibration heuristic based on persistence of 1-dimensional cycles.
CNNs reconstruct medium properties from wave probing responses.
problem Determining medium properties from wave responses.
method Deep convolutional neural networks (CNNs) for nonlinear wave equations.
result Quantitative dependence of network depth and units on medium complexity.
Deep neural network estimates support of sparse signals for improved phase retrieval.
problem Sparse phase retrieval from Fourier magnitudes with support estimation.
method Trained deep neural network (DNN) provides extended support estimate E larger than the support T. result DNN-based support estimation improves signal reconstruction performance with lower complexity.