End-to-end network predicts and aligns continuous emotion labels.
problem Inconsistent alignment of continuous emotion labels with speech signals.
method Convolutional neural network with a multi-delay sinc layer.
result State-of-the-art results in predicting and aligning continuous emotion labels.
Generative flows are attractive because they admit exact likelihood optimization and efficient image synthesis. Recently, Kingma & Dhariwal (2018) demonstrated with Glow that generative flows are capable of generating high quality images. We generalize the 1 x 1 convolutions proposed in Glow to invertible d x d convolu…
Spectral graph convolutional neural networks (CNNs) require approximation to the convolution to alleviate the computational complexity, resulting in performance loss. This paper proposes the topology adaptive graph convolutional network (TAGCN), a novel graph convolutional network defined in the vertex domain. We provi…
Convolutional autoencoders improve personalized recommendations from image-based data.
problem Lack of personalized recommendations in gastronomic platforms using image data.
method Used convolutional autoencoders to extract features from images and improve personalized recommendations.
result Convolutional autoencoders outperform standard deep features in image-based personalized recommendation systems.
New model forecasts VaR using quantile CNNs.
problem Forecasting Value at Risk (VaR) for different asset quantiles.
method Quantile Convolutional Neural Networks (qCNNs) for VaR forecasting.
result The qCNN model produces accurate VaR forecasts from asset price history.
The paper establishes conditions for strict power concavity in convolutions.
problem Conditions for strict power concavity in convolutions.
method Analyzes sufficient conditions for strict parabolic power concavity of convolutions.
result Establishes sufficient conditions for strict power concavity of convolutions.
Combines CNN and LSTM for spatio-temporal graph networks.
problem Improving spatio-temporal feature extraction.
method Proposes a new architecture combining CNN and LSTM temporal blocks.
result Empirical comparison shows our model outperforms existing models.
RecNets use RNNs to process image channels in a compact, recurrent way.
problem Creating efficient neural network architectures for computer vision.
method Introducing RecNets with CRC layers that simulate recurrent processing of image channels.
result RecNets achieve superior size-accuracy trade-off compared to other compact models.
In this work we detail the application of a fast convolution algorithm computing high dimensional integrals to the context of multiplicative noise stochastic processes. The algorithm provides a numerical solution to the problem of characterizing conditional probability density functions at arbitrary time, and we applie…
Paper analyzes the free energy of CNNs with skip connections in Bayesian learning.
problem Dependency of CNNs with skip connections on the number of parameters.
method Examines the Bayesian free energy of CNNs with and without skip connections.
result The upper bound of free energy of Bayesian CNN with skip connections does not depend on overparametrization.
Enhances speech emotion recognition by adapting to varying time scales.
problem Robust emotion recognition from speech audio with temporal variations.
method Introduces multi-time-scale (MTS) convolutional layers to CNNs.
result MTS layers improve generalization, especially on smaller datasets.
Mutual teaching improves graph models with less labeled data.
problem Training graph models with limited labeled data.
method Dual model training with mutual teaching strategy.
result Significant performance improvement with less labeled data.
Predicting diagnoses from Electronic Health Records (EHRs) is an important medical application of multi-label learning. We propose a convolutional residual model for multi-label classification from doctor notes in EHR data. A given patient may have multiple diagnoses, and therefore multi-label learning is required. We …
Generative networks improve fluid simulation quality by focusing on high frequencies.
problem Low-frequency details missing in fluid simulation reconstructions.
method Frequency-aware loss function for generative networks.
result Improved perceptual quality of fluid simulation results in mid-frequency bands.
Hybrid LSTM-fully convolutional networks (LSTM-FCN) for time series classification have produced state-of-the-art classification results on univariate time series. We show that replacing the LSTM with a gated recurrent unit (GRU) to create a GRU-fully convolutional network hybrid model (GRU-FCN) can offer even better p…
Enhanced deep CNNs improve cardiac abnormality diagnosis from ECGs.
problem Diagnosing cardiac abnormalities from 12-lead ECGs.
method Training an enhanced deep convolutional neural network with hand-crafted features, data preprocessing, and augmentation.
result Promising generalization performance in ECG diagnosis.
Proposes a new CNN for meshes that can handle orientation.
problem Isotropic kernels in graph convolutions are insensitive to mesh geometry.
method Introduces gauge equivariant kernels and geometric message passing.
result Significantly improved expressivity over conventional GCNs.
Bayesian convolutional deep sets improve ambiguity in stationary process modeling.
problem Ambiguity in translation equivariant functional representations due to insufficient data points.
method Introduce Bayesian convolutional deep sets with task-dependent stationary prior.
result Improves representation quality compared to kernel smoother and non-parametric models.
Optimizes CNN architectures by analyzing receptive fields without training.
problem CNNs often have too many layers, making them resource-intensive and less effective.
method Layer-wise analysis of receptive fields to identify unproductive layers.
result Identifies and removes unproductive layers, improving CNN performance and efficiency.
HKConv learns hyperbolic features by aggregating kernel points.
problem Challenges in learning good hyperbolic representations using Euclidean operations.
method Proposes HKConv, a trainable hyperbolic convolution that correlates local features with kernel points and aggregates them.
result HKConv learns expressive local features according to hyperbolic geometry and enjoys equivariance to permutation and invariance to parallel transport.
This paper proposes a new Quantum Spatial Graph Convolutional Neural Network (QSGCNN) model that can directly learn a classification function for graphs of arbitrary sizes. Unlike state-of-the-art Graph Convolutional Neural Network (GCNN) models, the proposed QSGCNN model incorporates the process of identifying transit…
This paper removes the finite variance assumption for deep convolutional neural networks.
problem Removing the finite variance assumption for deep convolutional neural networks.
method Assuming iid parameters distributed according to a stable distribution, the paper shows that the infinite-channel limit of a deep feed-forward convolutional neural network is a multivariate stable stochastic process.
result The infinite-channel limit of a deep feed-forward convolutional neural network, under suitable scaling, is a multivariate stable stochastic process.
Recurrent Neural Networks (RNNs) with sophisticated units that implement a gating mechanism have emerged as powerful technique for modeling sequential signals such as speech or electroencephalography (EEG). The latter is the focus on this paper. A significant big data resource, known as the TUH EEG Corpus (TUEEG), has …
Compressing convolutional neural networks (CNNs) is essential for transferring the success of CNNs to a wide variety of applications to mobile devices. In contrast to directly recognizing subtle weights or filters as redundant in a given CNN, this paper presents an evolutionary method to automatically eliminate redunda…
Proposes a new layer for efficient 3D shape discrimination.
problem Irregular structure and redundancy in 3D point clouds hinder efficient inter-class discrimination.
method Integrates Blended Convolution and Synthesis layer that projects and synthesizes 3D point clouds, followed by 3D convolution in the unit ball.
result End-to-end architecture achieves compelling results on 3D shape recognition and retrieval.
Superior performance and ease of implementation have fostered the adoption of Convolutional Neural Networks (CNNs) for a wide array of inference and reconstruction tasks. CNNs implement three basic blocks: convolution, pooling and pointwise nonlinearity. Since the two first operations are well-defined only on regular-s…
This paper reports the method and evaluation results of MedAusbild team for ISIC challenge task. Since early 2017, our team has worked on melanoma classification [1][6], and has employed deep learning since beginning of 2018 [7]. Deep learning helps researchers absolutely to treat and detect diseases by analyzing medic…
Graph scattering transforms are stable to metric perturbations of network topology.
problem Stability of graph data representations under metric perturbations.
method Extending scattering transforms to network data using multiresolution graph wavelets and graph convolutions.
result Graph scattering transforms are stable to metric perturbations of the underlying network topology.
Vector-valued neural learning has emerged as a promising direction in deep learning recently. Traditionally, training data for neural networks (NNs) are formulated as a vector of scalars; however, its performance may not be optimal since associations among adjacent scalars are not modeled. In this paper, we propose a n…
TTW aligns time-series faster and more accurately than existing methods.
problem Efficiently aligning multiple time-series signals with varying lengths.
method TTW uses a sinc convolutional kernel and gradient-based optimization for linear time and sequence complexity.
result TTW outperforms existing methods in time-series averaging and classification tasks.
The paper applies Information Bottleneck theory to CNNs and finds compression phase not always present.
problem Understanding the behavior of convolutional neural networks.
method Employed Information Bottleneck theory to analyze CNNs.
result Compression phase not observed in all CNN cases.
Two deep learning models improve indoor location prediction from WiFi fingerprints.
problem Indoor location prediction from WiFi fingerprints.
method Convolutional mixture density recurrent neural network and VAE-based semi-supervised learning model.
result Proposed models outperform existing methods in real-world datasets.
AEGCN uses autoencoder constraints to improve graph node classification.
problem Node classification on graph domains with reduced information loss.
method Autoencoder-constrained graph convolutional network (AEGCN).
result Adding autoencoder constraints significantly improves graph convolutional network performance.
Sparse molecular representations improve interpretability in graph neural networks.
problem Difficulty in understanding which molecular graph aspects drive deep learning predictions.
method Constrain weights in a graph convolutional neural network using the Gini index to maximize representation inequality.
result The Gini-constrained approach does not degrade evaluation metrics and allows for interpretable representation combination.
Graph neural networks can be adapted to new graphs with a limit object called graphon NNs.
problem Transferability of graph neural networks across different graphs.
method Introduced graphon NNs as limit objects of GNNs and proved a bound on the difference between GNN and graphon-NN outputs.
result The bound on the difference between GNN and graphon-NN outputs vanishes with growing number of nodes if the graph convolutional filters are bandlimited.
Human Activity Recognition (HAR) based on motion sensors has drawn a lot of attention over the last few years, since perceiving the human status enables context-aware applications to adapt their services on users' needs. However, motion sensor fusion and feature extraction have not reached their full potentials, remain…
A new method for comparing image probability measures using convolution operators.
problem Efficiently comparing images using conventional sliced Wasserstein methods.
method Proposed convolution sliced Wasserstein (CSW) methods with stride, dilation, and non-linear activation.
result CSW demonstrates favorable performance over conventional sliced Wasserstein in image comparison and deep generative modeling.
This paper uses graph convolutional networks to improve the accuracy of neural architecture search.
problem Improving the precision of sampled sub-networks in weight-sharing NAS.
method Training a graph convolutional network to fit the performance of sampled sub-networks.
result Achieved higher rank correlation coefficient and better final architecture performance.
Graph convolutional networks refine organ segmentation using uncertainty analysis.
problem Challenges in organ segmentation due to variability and tissue similarity.
method Uncertainty analysis of graph convolutional networks for semi-supervised learning.
result Improved segmentation accuracy (1% for pancreas, 2% for spleen) compared to state-of-the-art methods.
Max-pooling architectures are theoretically analyzed and shown to be globally optimized and generalize well.
problem Theoretical understanding and optimization of max-pooling in deep learning architectures.
method Theoretical analysis of a convolutional max-pooling architecture, focusing on a pattern detection problem.
result Max-pooling architectures can be globally optimized and generalize well, even for highly over-parameterized models.
Convolutional Neural Network is known as ConvNet have been extensively used in many complex machine learning tasks. However, hyperparameters optimization is one of a crucial step in developing ConvNet architectures, since the accuracy and performance are reliant on the hyperparameters. This multilayered architecture pa…
Proposes ECLSTM for more accurate RUL estimation from time series data.
problem Predicting Remaining Useful Life (RUL) from multivariate time series data.
method Embedded Convolutional LSTM (ECLSTM) with automated hyperparameter optimization.
result ECLSTM outperforms state-of-the-art approaches on benchmark data sets.
We present a new convolutional neural network-based time-series model. Typical convolutional neural network (CNN) architectures rely on the use of max-pooling operators in between layers, which leads to reduced resolution at the top layers. Instead, in this work we consider a fully convolutional network (FCN) architect…
Improves Gaussian process models for large datasets.
problem Selecting proper covariance functions in Gaussian processes.
method Nonparametric process convolutions and deep GP models.
result Improves performance on benchmarks for GPs, especially for larger datasets.
Absum improves CNN robustness against Fourier attacks.
problem Structural sensitivity of CNNs to Fourier basis functions.
method Regularizes convolution filter weights to reduce sensitivity.
result Improves robustness against single Fourier attack.
A new method compresses point clouds efficiently, outperforming existing techniques.
problem Efficiently compressing large point cloud datasets for VR applications.
method Learned convolutional transforms and uniform quantization for joint rate and distortion optimization.
result Significant rate-distortion improvement (51.5% BDBR savings) on Microsoft Voxelized Upper Bodies dataset.
Improved Zero learning framework wins games like Hex and Havannah.
problem Improving Zero learning for board games with limited resources.
method Fully convolutional structure, global pooling, and checkpointing.
result Won against strong humans in Hex and Havannah.
A convolution neural network (CNN) based classification method for broadband DOA estimation is proposed, where the phase component of the short-time Fourier transform coefficients of the received microphone signals are directly fed into the CNN and the features required for DOA estimation are learnt during training. Si…