This work prunes CNN filters based on their functionality, not just size.
problem Redundant filters in CNNs waste computation resources.
method Functionality-oriented filter pruning method.
result Pruning based on functionality optimizes computation and interprets filter importance.
Paper improves graph convolutional networks by adjusting filter size.
problem Improving predictive performance of graph convolutional networks.
method Introducing a hyper-parameter to influence filter size in graph convolutions.
result Improves predictive performance of Deep Graph Convolutional Networks.
Large filters improve performance but are costly; this work uses learned box filters and summed-area tables.
problem Improving performance in dense prediction tasks like human pose estimation with large filters.
method Adopted learnable box filters and summed-area tables to reduce computational cost and maintain performance.
result Demonstrated competitive performance on human pose estimation benchmarks.
We generate transformation-invariant CNNs using context-aware filters.
problem Creating transformation-invariant neural networks for image recognition.
method Input-conditioned convolution filters combined with max-pooling and multi-instance learning.
result Significantly improved error rates on MNIST variations (1.13% on MNIST-rot-12k, 1.12% on Half-rotated MNIST, 0.68% on Scaling MNIST).
Convolutional Bayesian filtering generalizes state estimation by incorporating inequality conditions.
problem Standard Bayesian filtering assumes exact conditional probabilities, limiting its applicability.
method Introducing inequality conditions transforms conditional probabilities into convolutional forms, expanding the filtering framework.
result Convolutional Bayesian filtering encompasses standard Bayesian filtering and allows for more nuanced model consideration.
BankGCN improves graph convolution networks by handling multi-channel signals with adaptive filter banks.
problem Handling multi-channel graph signals with limited architectures.
method BankGCN decomposes multi-channel signals into subspaces and uses adapted filters for each subspace.
result BankGCN achieves excellent performance in graph classification on benchmark datasets.
Paper presents efficient algorithms for convolutional neural networks using Winograd minimal filtering.
problem Resource-efficient implementation of convolutional neural networks.
method Winograd minimal filtering trick applied to M-tap filters (M=3,5,7,9,11) for parallel hardware implementation.
result Approximately 30% reduction in multipliers for fully parallel hardware implementation.
Paper learns to rotate filters for group convolutions.
problem Difficult to rotate 3x3 filters on pixel grids.
method Learn filter basis and rotation-invariant coefficients; switch basis for rotation.
result Produces feature maps insensitive to input rotations.
ARMA filters improve graph neural networks' flexibility and robustness.
problem Improving graph neural networks' performance and robustness.
method Proposes an ARMA-based graph convolutional layer for graph neural networks.
result ARMA filters outperform polynomial filters in various graph tasks.
DCFNet decomposes CNN filters into learned coefficients with bases, reducing parameters and computation.
problem Reduction of model parameters and computation in CNNs.
method DCFNet decomposes convolutional filters into a truncated expansion with pre-fixed bases, learning only the coefficients.
result DCFNet maintains accuracy for image classification tasks with significantly fewer parameters, including with random bases.
FSNet compresses CNNs by sharing weights in 1D vectors.
problem Compressing deep CNNs while maintaining performance.
method Learning a 1D vector (Filter Summary) to share weights in convolutional filters.
result FSNet reduces parameter space and accelerates computation.
The paper analyzes how gradient descent learns convolutional filters for non-Gaussian inputs.
problem Learning convolutional filters with ReLU for non-Gaussian input distributions.
method Analysis of gradient descent convergence for ReLU activation with polynomial time complexity.
result Gradient descent can learn convolutional filters in polynomial time, with convergence rate dependent on input distribution smoothness and patch similarity.
Evolutionary method reduces CNN complexity for mobile devices.
problem Efficiently compressing CNNs for mobile devices.
method Evolutionary algorithm to identify and remove redundant convolution filters.
result Generates an extremely compact CNN with improved compression and speed-up ratios.
Paper introduces context-sensitive filters for better text processing.
problem Static filters limit NLP performance; need dynamic context.
method Meta network learns context-aware filters for sentences.
result Context-sensitive filters improve NLP tasks.
RotDCF decomposes CNN filters for rotation-equivariant deep networks.
problem Handling global deformations in images for vision tasks.
method Decomposes convolutional filters over joint steerable bases for rotation-equivariance.
result Significantly reduces model size and computational complexity while preserving performance.
Paper simplifies CNNs for irregular data using MIMO graph filters.
problem Challenges in applying CNNs to irregularly structured data.
method Introduces MIMO graph filters to CNNs, simplifying architectures.
result Proposed architectures reduce model complexity and computational cost.
New PTC convolution preserves properties of Euclidean convolutions on manifolds.
problem Lack of generalizable convolutions on curved domains with desirable properties.
method Parallel transport convolution (PTC) on Riemannian manifolds.
result PTC preserves compactly supported filters and directionality.
HexaConv improves CNN performance by using hexagonal filters and group convolutions.
problem Improving CNN performance by exploiting more symmetries.
method Implementing planar and group convolutions over hexagonal lattices.
result HexaConv outperforms conventional CNNs on aerial scene classification.
Improved GCNs for non-sparse graphs with low-rank filters.
problem Training and evaluation of GCNs on large non-sparse graphs is computationally expensive.
method Introduced low-rank filters and a reduced-order GCN architecture.
result Significant runtime acceleration and improved accuracy achieved.
Novel CNN integrates learnable FIR filters for heart sound detection.
problem Automatic detection of heart sound abnormalities for early diagnosis.
method Proposes a CNN with tConv layers to learn FIR filter-bank parameters.
result Proposed models outperform state-of-the-art systems in heart sound detection.
The paper bounds the complexity of GCNs using Rademacher complexity.
problem Understanding the sample complexity of GCNs.
method Derived tight upper and lower bounds of Rademacher complexity for GCN models.
result The derived bounds depend on the largest eigenvalue of the graph filter and the degree distribution.
Centripetal SGD prunes deep CNNs by making filters collapse.
problem Pruning deep CNNs with complex structures.
method Centripetal SGD, a novel optimization method.
result Pruning deep CNNs without performance loss.
The study reveals flaws in pruning criteria and proposes a new assumption for better filter selection.
problem Flaws in existing pruning criteria for CNNs.
method Empirical experiments and Convolutional Weight Distribution Assumption.
result The Convolutional Weight Distribution Assumption improves filter selection in pruning.
New filter bank regularization improves DCNNs by incorporating image priors.
problem Improving DCNNs' robustness and generality.
method Structured filter bank regularization of DCNN kernels.
result Filter bank regularization leads to faster convergence and better generalization.
Developed a framework for designing filters in spectral GCNNs with improved performance.
problem Designing effective filters for spectral GCNNs with regularization properties.
method Exploring regularization properties of graph Laplacian and proposing a generalized framework for filter design.
result New filters derived from the framework outperform state-of-the-art techniques in semi-supervised node classification.
ScDCFNet improves multiscale image classification with reduced model size.
problem Improving performance in multiscale image classification.
method Decomposed convolutional filters for ST-equivariant CNNs.
result ScDCFNet achieves significantly improved performance in multiscale image classification.
Convolutional neural networks learn phase-dependent frequency representations.
problem Capturing phase dependence in frequency representations for better signal analysis.
method Convolutional neural networks learn filters with different phases, which rectify to phase-dependent descriptors.
result Phase harmonics correlations can compressively represent signals with sparse wavelet coefficients.
This work ranks CNN filters based on their importance.
problem Unclear role of CNN neurons in producing output.
method Two methods: Shapley value game theory and Importance switch variational inference.
result Filters with higher importance are more crucial for output.
Enhances DNNs by selectively learning key image edges.
problem Improving DNN robustness against adversarial attacks.
method Introduces Secure Selective Convolution (SSC) to learn important image edges.
result Significant reduction in attack success rate and imperceptibility of adversarial images.
CAOL learns filters from large datasets, and this paper analyzes its performance.
problem Understanding how dataset size impacts CAOL filter updates.
method Analyzes the impact of dataset size on CAOL filter updates using deterministic and probabilistic bounds.
result Empirical evidence suggests that using more training data can improve CAOL performance.
A new method to compress CNNs without changing architectures.
problem Redundancy in CNN filter weights.
method Replacing filters with a fixed set and a learnable 1D layer.
result Significant reduction in parameters and operations with minimal performance loss.
New CAOL framework learns diverse convolutional filters for improved signal recovery.
problem Memory limitations in patch-domain approaches for learning kernels from large datasets.
method Convolutional Analysis Operator Learning (CAOL) framework with BPEG-M method.
result CAOL significantly accelerates convergence and improves reconstruction quality.
New graph convolution captures local features on non-Euclidean grids.
problem Capturing local features on irregular, coarse non-Euclidean grids.
method Low-rank learnable local filters in graph convolutions.
result Proves more expressive than previous spectral graph convolution methods.
GCN improved for large graphs with LCF to reduce complexity and noise.
problem Efficiency and effectiveness of graph convolution in large graphs.
method Proposed Low-pass Collaborative Filter (LCF) to simplify graph convolution.
result Significant improvement in effectiveness and efficiency of GCN.
TAGCN improves graph CNN performance without approximation.
problem Performance loss in spectral graph convolutional neural networks.
method Topology adaptive graph convolutional network (TAGCN) with adaptive filters.
result TAGCN outperforms existing spectral CNNs on various datasets.
A new neural network separates singing voices more effectively.
problem Separating singing voices from mixed signals with high accuracy.
method MBR-FCN that processes different frequency bands with varying resolutions and filters.
result The MBR-FCN achieves better performance with fewer parameters.
The paper addresses instability in CNNs' first layer by proving max pooling's shift invariance.
problem Instability in CNNs' first layer, leading to sensitivity to small input shifts.
method Establishing conditions for max pooling's shift invariance and deriving a measure of stability.
result Max pooling approximates a nearly shift-invariant complex modulus under certain conditions.
RGCF improves collaborative filtering by refining graph convolution embeddings.
problem GCN-based recommendation models introduce noise and redundancy, limiting high-order connectivity capture.
method Developed RGCF, a new GCN-based Collaborative Filtering model with redesigned embeddings.
result RGCF significantly outperforms state-of-the-art models on public datasets.
Paper proves spectral filters can be transferred between graphs.
problem Proving spectral filters can be transferred between graphs.
method Introducing the Cayley smoothness space and proving filters in this space are linearly stable.
result Graph spectral filters are transferable if they are in the Cayley smoothness space.
ConvTimeNet is a pre-trained CNN for time series classification.
problem Training deep neural networks for time series classification requires careful tuning and resources.
method ConvTimeNet is a pre-trained deep convolutional neural network trained on diverse univariate time series datasets. It adapts to new tasks with minimal fine-tuning.
result ConvTimeNet achieves significant gains in classification accuracy and computational efficiency compared to existing methods.
Stable algebraic filters improve neural network performance.
problem Improving neural network stability to deformations.
method Analyzed stability of algebraic filters and neural networks under deformations of the homomorphism.
result Stable algebraic filters have frequency responses whose derivative is inversely proportional to frequency.
Visualizes deep neural networks for speech recognition using learned topographic filter maps.
problem Unintuitive internal structure of deep neural networks complicates activation visualization.
method Trains a convolutional speech recognition model with filters arranged in a 2D grid, highlighting similar filters.
result Topographic filter maps visualize artificial neuron activations more intuitively.
Prunes CNNs by removing redundant filters with provable guarantees.
problem Redundant filters in over-parameterized neural networks.
method Sampling-based approach using saliency scores and importance sampling.
result Consistently generates sparser and more efficient models.
Proposes interpretable filters in CNNs for object classification.
problem Learning interpretable filters in deep CNNs without additional annotations.
method Assigns each filter in a CNN to an object part during training.
result Interpretable filters are more semantically meaningful than traditional filters.
CNN predicts stock fluctuations using company news headlines.
problem Predicting next-day stock fluctuations based on company-specific news.
method Convolutional Neural Network (CNN) with reduced filter dimensions and multiple hidden layers. Fine-tuned word embeddings and various filter widths.
result 61.7% classification accuracy achieved using pre-learned embeddings.
Paper proposes DeCEF layers to reduce CNN complexity.
problem Reduces complexity of CNNs without pre-trained models.
method Develops Depthwise Convolutional Eigen-Filter (DeCEF) layers.
result Achieves similar or higher accuracy with 2/3 parameters and 2/3 FLOPs.
GCRNNs improve graph problem solving with fewer parameters.
problem Graph process problems like earthquake epicenter identification and weather prediction.
method GCRNNs use convolutional filter banks and time-gated variations of GCRNNs (Gated GCRNNs) to improve performance.
result GCRNNs significantly improve performance over GNNs and another graph recurrent architecture.
SchNet models quantum interactions using continuous filters, outperforming traditional methods.
problem Capturing continuous atomic positions in molecules without losing physical information.
method Continuous-filter convolutional neural network architecture in SchNet.
result SchNet models both total energy and interatomic forces with rotationally invariant predictions and a smooth potential energy surface.