Generalizes PCA and ICA for continuous-time signals using neural networks.
problem Low-rank decomposition of continuous-time vector-valued signals.
method Implicit neural network framework to learn numerical approximations of PCA and ICA.
result Unified approach to PCA and ICA in continuous domain, enforcing decorrelation and independence.
Improved language identification accuracy through signal combination methods.
problem Enhancing speech recognition accuracy across multiple languages.
method Combining low-level acoustic signals with language-specific recognizer signals using lattice-based ensemble models and deep neural networks.
result Deep neural network model outperforms lattice-based ensemble model, reducing error rate from 5.5% to 4.3%.
Neural network detects spectrum signals under uncertain parameters.
problem Spectrum sensing under uncertain primary user signal parameters.
method Trained neural network to detect modulated signals robustly.
result Neural network gains robustness under carrier frequency, phase, and symbol time offsets.
DyEnsemble improves BCI accuracy by adapting to nonstationary neural signals.
problem Nonstationary neural signals in BCI cause decoding errors.
method Dynamic ensemble modeling that learns and combines diverse models online.
result DyEnsemble outperforms Kalman filters, especially with noisy signals.
Neural signals are characterized by rich temporal and spatiotemporal dynamics that reflect the organization of cortical networks. Theoretical research has shown how neural networks can operate at different dynamic ranges that correspond to specific types of information processing. Here we present a data analysis framew…
Light neural network detects modulation in noisy signals.
problem Efficiently detecting modulation in noisy signals.
method Light neural network architecture invariant to impairments.
result Network achieves accuracy under realistic impairments.
A new neural network captures and explains trajectory patterns.
problem Analyzing complex spatial trajectories in urban planning and neuroscience.
method Composite Signal Neural Networks (CompSNN) combining three interpretable ANN modules.
result CompSNN outperforms individual modules and visualizes useful signal parts.
Unified deep learning for graph signals, simplifying existing models.
problem Efficiency of Convolutional Neural Networks on graph signals.
method Unified formalism for existing deep learning models on graph signals.
result Unified formalism simplifies and compares existing models.
DeepCodec learns to take undersampled measurements and recover signals using deep neural networks.
problem Signal recovery from undersampled data.
method Adaptive deep convolutional neural networks for sensing and recovery.
result DeepCodec outperforms traditional ℓ1-minimization in signal recovery. The paper analyzes deep neural networks using rectified linear units.
problem Understanding the individual affine linear representations of deep neural networks.
method Signal processing perspective, atomic decompositions, Lipschitz regularity estimation.
result Conditions for stabilizing learning in deep neural networks without network depth constraints.
Bayesian priors improve neural network performance on weak signals.
problem Challenges in encoding domain knowledge for weak signals in neural networks.
method Proposed a new joint prior over local scale parameters for feature sparsity and signal-to-noise ratio, optimized with Stein gradient.
result Improved prediction accuracy on various datasets, including genetics applications with weak and sparse signals.
Deep neural networks improve angle of arrival estimation with lower complexity.
problem Estimating the number of sources and their angles of arrival from a single antenna array observation.
method Apply a deep neural network (DNN) approach to the problem.
result Deep neural networks can attain maximum likelihood performance with feasible complexity and outperform other methods.
A new neural network improves frequency estimation from noisy signals.
problem Estimating frequencies of sinusoidal components in noisy signals.
method A novel neural network architecture combined with a module to detect the number of frequencies.
result Significantly more accurate frequency estimation at medium-to-high noise levels.
CNNs improve signal-background classification in particle physics experiments.
problem Improving accuracy in classifying signal from background in particle physics experiments.
method Extensive convolutional neural architecture search for 2D and 3D image data.
result Achieved high accuracy for signal/background discrimination with CNNs, less parameters than ResNet.
Goldilocks activation functions improve neural network performance.
problem Improving neural network performance and understanding signal transformation.
method Introducing Goldilocks activation functions that locally deform input signals.
result Goldilocks networks outperform or match SELU and RELU on CIFAR-10 and CIFAR-100 datasets.
New findings suggest weight maps from classifiers may not reliably indicate neural signals.
problem The reliability of interpreting weight maps from classifiers in neuroimaging studies.
method Used semi-simulated ECoG data to investigate signal-to-noise ratio and sparsity effects.
result Not all cases produce false positives and high-weight features are unlikely to be FP.
Study eigenvalues and eigenvectors in neural networks, focusing on signal propagation.
problem Characterize signal eigenvalues and eigenvectors in neural networks.
method Characterizes signal eigenvalues and eigenvectors for a nonlinear spiked covariance model.
result Provides precise quantitative characterizations of signal eigenvalues and eigenvectors in neural networks.
A new neural network model uses polynomial chaos theory to improve neural signal processing.
problem Redundant neural signal representation in DANNs.
method Employing arbitrary polynomial chaos theory to construct orthonormal representations in DANNs.
result Improves neural signal processing by reducing redundancy and enhancing orthogonality.
TSN improves sparse signal recovery with less complexity.
problem Sparse regression problem of recovering sparse signals from measurements.
method Tree search algorithm driven by deep neural network with pruning.
result TSN outperforms conventional methods in various sensing matrices.
New neural network learns like humans without backpropagation.
problem Deep learning without backward error propagation.
method Biologically inspired feedforward supervisory signal.
result Effective learning from large amounts of feedforward information.
Paper introduces compressibility loss for learning sparse neural network weights.
problem Learning highly compressible neural network weights.
method Applying a compressibility loss to minimize the negated sparsity of the signal.
result At critical points, weight vectors are ternary signals with a sparsity directly related to the objective value.
Neural process model improves real-time condition monitoring signal prediction.
problem Real-time adaptation for complex condition monitoring signals.
method Label-aware neural processes encoding and reconstruction.
result Advantages in real-time adaptation, enhanced signal prediction with uncertainty quantification, and joint prediction for labels and signals.
This work uses encoder-decoder networks to denoise one-dimensional signals by aligning clean and noisy signal latent representations.
problem Noise removal in one-dimensional signals, especially in medical and motion signals.
method Encoder-decoder architecture with adversarial learning to align clean and noisy signal latent representations.
result Better performance on electrocardiogram and motion signal denoising compared to learning-based and non-learning approaches.
Two graph neural network architectures improve CNN performance for graph signals.
problem Improving CNN performance for graph signals.
method Introducing two graph neural network architectures: selection GNN and aggregation GNN.
result Multinode aggregation GNNs consistently perform best in source localization and authorship attribution tasks.
New RNN reconstructs video frames from sparse measurements.
problem Sequential signal reconstruction from compressive measurements.
method Unfolding proximal gradient method for l1-l1 minimization.
result Outperforms state-of-the-art RNN models in video frame reconstruction.
A neural network learns to estimate spectra from few noisy samples.
problem Estimating spectra from limited noisy data.
method Training a neural network on simulated data to approximate multisinusoidal signal spectra.
result The approach performs well in various noise conditions and is competitive with classical methods.
New method separates mixed distributions without requiring samples of each source.
problem Separating mixed distributions in machine learning and signal processing.
method Neural Egg Separation method iteratively learns to separate known from unknown distributions.
result Neural Egg Separation outperforms current methods in audio and image separation tasks.
FreSh shifts model's initial frequency spectrum to match target signal, improving neural representation performance.
problem MLPs' low-frequency bias limits capturing high-frequency details accurately.
method FreSh selects embedding hyperparameters to align model's initial output spectrum with target signal's spectrum.
result FreSh improves performance across various neural representation methods and tasks with minimal computational overhead.
Deep neural networks help recover two signals from noisy mixtures.
problem Recovering two signals from noisy subgaussian mixtures with prior structural information.
method Used deep generative neural networks (GNNs) to solve the demixing problem for Lipschitz signals.
result Proved a sample complexity bound for nearly optimal recovery error, extending previous results.
New method shows random, diverse initializations are not essential for deep neural networks.
problem The necessity of random, diverse initializations in deep neural networks.
method Constructed a deep convolutional network with identical features by initializing weights to 0, enabling signal propagation and stable gradients.
result Random, diverse initializations are not necessary for training neural networks.
Graph neural networks enhance IceCube neutrino detection.
problem Improving signal detection in IceCube neutrino observatory.
method Leveraged graph neural networks to model IceCube detector array as a graph, with vertices as sensors and edges based on spatial coordinates.
result GNN outperforms traditional methods in classifying IceCube events.
New framework explains deep learning using signal processing techniques.
problem Lack of a rigorous mathematical theory explaining deep learning performance.
method Transform-domain sparse regularization, Radon transform, and approximation theory.
result Explains neural network properties and performance.
VPNet uses variable projection for efficient neural network training.
problem Efficient and interpretable neural network training for signal processing.
method Variable projection (VP) applied to neural networks.
result VPNet achieves fast learning and good accuracy with low computational cost.
GEM learns a manifold for cross-modal data, capturing structure without modality dependence.
problem Modality-specific neural models limit flexibility and custom architecture.
method Casts learning as manifold inference, enforcing coverage, linearity, and isometry.
result GEM learns latent structure across image, shape, audio, and cross-modal domains.
Improved jet tagging reduces systematic uncertainties and enhances signal purity.
problem Boosted resonance decay signals from jets are difficult to distinguish from background.
method Adversarial neural networks to decorrelate jet substructure tagger.
result Adversarial trained tagger outperforms conventional methods in discovery significance.
New method labels EEG recordings for efficient neural decoding evaluation.
problem Challenges in evaluating neural decoding methods on limited, noisy data.
method Post-hoc labeling of arbitrary EEG recordings for generating labeled datasets.
result Generates large labeled datasets for benchmarking neural decoding methods.
Attention mechanism combines bottom-up and top-down signals in neural networks.
problem Combining robust perception with bottom-up and top-down signals.
method Attention mechanism over modulated recurrent neural networks.
result Bidirectional information flow leads to improved performance in various tasks.
CNN detects multipath in GNSS signals using correlator output.
problem Multipath contamination errors in GNSS signals.
method Deep Convolutional Neural Networks (CNN) applied to GNSS correlator output.
result CNN accurately detects multipath in GNSS signals.
Stabilizes deep Bayesian neural networks with self-stabilizing priors.
problem Brittleness and difficulty in training deep Bayesian neural networks.
method Signal propagation theory, reformulated ELBO, self-stabilizing priors.
result Improved convergence and robustness in training deeper networks and noisier settings.
STFNets learn signals from time-frequency perspective, outperforming traditional models.
problem Learning signals from IoT data with better features in the frequency domain.
method Integrates Short-Time Fourier Transform into neural networks for direct frequency domain feature learning.
result Significantly outperforms state-of-the-art models in various experiments.
MCLNN improves audio classification with binary masks.
problem Improving audio classification accuracy.
method Binary mask applied to CLNN for feature preservation and combination exploration.
result Competitive recognition accuracies on GTZAN and ISMIR2004 datasets.
Improved Schizophrenia diagnosis using brain signal features with limited observations.
problem Ambulatory diagnoses of neuronal diseases with limited brain signal data.
method Pairwise distance learning approach using Siamese neural network and cosine contrastive loss.
result Improved accuracy and sensitivity in Schizophrenia diagnosis (+10pp).
In this paper, we propose majority voting neural networks for sparse signal recovery in binary compressed sensing. The majority voting neural network is composed of several independently trained feedforward neural networks employing the sigmoid function as an activation function. Our empirical study shows that a choice…
New LSTM variants reduce complexity and parameters without sacrificing performance.
problem Complexity and large parameters in LSTM networks.
method Eliminating combinations of gating signals to reduce parameters.
result Three new LSTM variants achieve comparable performance to standard LSTM with fewer parameters.
New model resolves signal ambiguities in ill-posed systems.
problem Signal retrieval from indirect measurements with known models.
method Variational generative model that captures signal distribution.
result Retrieves consistent signals with high fidelity.
Deep neural network detects driver intentions from video.
problem Detecting driver intentions for safer self-driving.
method Uses deep learning to analyze turn signals and emergency flashers.
result High per-frame accuracy in challenging scenarios.
Deep neural networks predict walking, biking, and driving from Wi-Fi signals.
problem Predicting human mobility modes using Wi-Fi signals.
method Deployed Wi-Fi sensors at four locations, developed and tested multiple classifiers (MLP, Decision Tree, Bagged Decision Tree, Random Forest).
result Multilayer Perceptron achieved 86.52% correct predictions of mobility modes.
Paper enhances haptic signals distinguishability with boosted technique.
problem Lack of large datasets in haptics domain limits feature extraction.
method General framework for haptic signal analysis, using spectral features and boosted embedding.
result Framework needs less training data and outperforms state-of-the-art.