Scattering Transforms (or ScatterNets) introduced by Mallat in [1] are a promising start into creating a well-defined feature extractor to use for pattern recognition and image classification tasks. They are of partic...
详细信息
The proceedings contain 91 papers. The topics discussed include: online function minimization with convex random relu expansions;text to image generative model using constrained embedding space mapping;identification ...
ISBN:
(纸本)9781509063413
The proceedings contain 91 papers. The topics discussed include: online function minimization with convex random relu expansions;text to image generative model using constrained embedding space mapping;identification of a thermal building model by learning the dynamics of the solar flux;mutual singular spectrum analysis for bioacoustics classification;Hankel subspace method for efficient gesture representation;texture classification from single uncalibrated images: random matrix theory approach;Gaussian density guided deep neural network for single-channel speech enhancement;and a recurrent encoder-decoder approach with skip-filtering connections for monaural singing voice separation.
The proceedings contain 106 papers. The topics discussed include: stochastic unfolding;iterative collaborative filtering for recommender systems with sparse data;learning with the kernel signal to noise ratio;semisupe...
ISBN:
(纸本)9781467310260
The proceedings contain 106 papers. The topics discussed include: stochastic unfolding;iterative collaborative filtering for recommender systems with sparse data;learning with the kernel signal to noise ratio;semisupervised kernel orthonormalized partial least squares;differential edit distance as a countermeasure to video scene ambiguity;haussdorff and hellinger for colorimetric sensor array classification;optimal cost function and magnitude power for NMF-based speech separation and music interpolation;a subspace learning algorithm for microwave scattering signal classification with application to wood quality assessment;level sets for retinal vasculature segmentation using seeds from ridges and edges from phase maps;breast ultrasound images gland segmentation;on the generalization ability of distributed online learners;and online learning for quality-driven unequal protection of scalable video.
The proceedings contain 73 papers. The topics discussed include: discriminating bipolar disorder from major depression using whole-brain functional connectivity: a feature selection analysis with SVM-FOBA algorithm;ch...
ISBN:
(纸本)9781467374545
The proceedings contain 73 papers. The topics discussed include: discriminating bipolar disorder from major depression using whole-brain functional connectivity: a feature selection analysis with SVM-FOBA algorithm;characterization of diabetic peripheral neuropathy in infrared video sequences using independent component analysis;active learning for efficient querying from a human oracle with noisy response in a language model assisted brain computer interface;maximum a posteriori estimation of relative abundances of protein conformations;synthetic structural magnetic resonance image generator improves deep learning prediction of schizophrenia;multiview classification of brain data through tensor factorization;an efficient multi-resolution SVM network approach for object detection in aerial images;and semi-supervised color decomposition for histopathological images using exclusive component analysis.
The proceedings contain 102 papers. The topics discussed include: non-negative matrix completion for bandwidth extension: a convex optimization approach;diffusion map for clustering FMRI spatial maps extracted by inde...
ISBN:
(纸本)9781479911806
The proceedings contain 102 papers. The topics discussed include: non-negative matrix completion for bandwidth extension: a convex optimization approach;diffusion map for clustering FMRI spatial maps extracted by independent component analysis;segmentation of medical images based on hierarchical evolutionary and bee algorithms;the linear process mixture model;video-informed approach for enhancing audio source separation through noise source suppression;introducing a simple fusion framework for audio source separation;orthogonal segmented model for underdetermined blind identification and separation of sources with sparse events;a constrained approach for extraction of pre-ICTAL discharges from scalp EEG;complex support vector machines for quaternary classification;MOGT: oversampling with a parsimonious mixture of Gaussian trees model for imbalanced time-series classification;and non-linear noise adaptive Kalman filtering via variational Bayes.
The proceedings contain 102 papers. The topics discussed include: a block-based combined scheme exploiting sparsity in nonlinear acoustic echo cancellation;Gaussian processes for music audio modeling and content analy...
ISBN:
(纸本)9781509007462
The proceedings contain 102 papers. The topics discussed include: a block-based combined scheme exploiting sparsity in nonlinear acoustic echo cancellation;Gaussian processes for music audio modeling and content analysis;enhanced distance subset approximation using class-specific subspace kernel representation for kernel approximation;approximate state-space Gaussian processes via spectral transformation;articulatory and spectrum features integration using generalized distillation framework;speech intelligibility prediction as a classification problem;detecting trends in twitter time series;speaker identification and clustering using convolutional neural networks;a neural network based algorithm for speaker localization in a multi-room environment;and mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorization with beta-divergence.
The proceedings contain 98 papers. The topics discussed include: hierarchical graph neural nets can capture long-range interactions;self-attention for audio super-resolution;deep complex convolutional recurrent networ...
ISBN:
(纸本)9781728163383
The proceedings contain 98 papers. The topics discussed include: hierarchical graph neural nets can capture long-range interactions;self-attention for audio super-resolution;deep complex convolutional recurrent network for multi-channel speech enhancement and dereverberation;model selection of kernel ridge regression for extrapolation;optimizing time domain fully convolutional networks for 3D speech enhancement in a reverberant environment using perceptual losses;online DOA estimation for noninteger linear antenna arrays in coarray domain;tracking of quantized signals based on online kernel regression;and robustness-aware filter pruning for robust neural networks against adversarial attacks.
The proceedings contain 98 papers. The topics discussed include: optimal pricing in black box producer-consumer Stackelberg games using revealed preference feedback;learning warm-start points for AC optimal power flow...
ISBN:
(纸本)9781728108247
The proceedings contain 98 papers. The topics discussed include: optimal pricing in black box producer-consumer Stackelberg games using revealed preference feedback;learning warm-start points for AC optimal power flow;minimax active learning via minimal model capacity;multi-step chord sequence prediction based on aggregated multi-scale encoder-decoder networks;robust hybrid beamforming with quantized deep neural networks;a machinelearning approach for classifying faults in microgrids using wavelet decomposition;robust importance-weighted cross-validation under sample selection bias;interpretable online banking fraud detection based on hierarchical attention mechanism;and a benchmark study of backdoor data poisoning defenses for deep neural network classifiers and a novel defense.
The proceedings contain 89 papers. The topics discussed include: mahalanobis-based one-class classification;improving the robustness of surface enhanced Raman spectroscopy based sensors by Bayesian non-negative matrix...
ISBN:
(纸本)9781479936946
The proceedings contain 89 papers. The topics discussed include: mahalanobis-based one-class classification;improving the robustness of surface enhanced Raman spectroscopy based sensors by Bayesian non-negative matrix factorization;data mining by nonnegative tensor approximation;non-negative tensor factorization with missing data for the modeling of gene expressions in the human brain;multiple speaker tracking with the factorial von mises-fisher filter;a probabilistic approach to hearing loss compensation;coherent time modeling of semi-Markov models with application to real-time audio-to-score alignment;ultra-low-power voice-activity-detector through context and resource-cost-aware feature selection in decision trees;and a probabilistic approach for phase estimation in single-channel speech enhancement using von mises phase priors.
The proceedings contain 121 papers. The topics discussed include: composer style-specific symbolic music generation using vector quantized discrete diffusion models;sparsification of deep neural networks via ternary q...
ISBN:
(纸本)9798350372250
The proceedings contain 121 papers. The topics discussed include: composer style-specific symbolic music generation using vector quantized discrete diffusion models;sparsification of deep neural networks via ternary quantization;orthogonal symmetric nonnegative matrix tri-factorization;convexity based pruning of speech representation models;discriminative community detection for multiplex networks;novel gradient sparsification algorithm via Bayesian inference;adaptive semantic image transmission using generative foundation model;FOLEYGEN: visually-guided audio generation;graph-vector autorregressive model for lithium-ion cell capacity estimation;and episodic fine-tuning prototypical networks for optimization-based few-shot learning: application to audio classification.
暂无评论