Paper learns an explainer to interpret CNN features without annotations.
problem Interpreting complex features in CNNs without labeled data.
method Unsupervised learning of an explainer to decompose and reconstruct feature maps.
result Explainer learns to reconstruct CNN features without losing information.
Enhanced financial reward with shuffled feature CNN-DRL.
problem Improving reward in financial data using CNN-DRL.
method Applying shuffled features to financial data for CNN-DRL.
result Substantial enhancement in reward attainment.
Measures CNN features' discriminative power for various datasets.
problem Improving feature representation transfer in CNNs.
method Statistical analysis of CNN features across 11 datasets.
result Low and middle level features behave differently under certain conditions.
3D-CNN method visualizes localized geometric features for manufacturability analysis.
problem Interpreting 3D-CNN decisions for complex geometries.
method 3D-CNN with surface normals, 3D-GradCAM for feature visualization.
result Identifies critical local features for manufacturability.
Proposes deep multimodal fusion for biometric identification.
problem Improving biometric identification accuracy with multiple modalities.
method Joint optimization of multiple modality-specific CNNs at different feature abstraction levels.
result Significant improvement in multimodal person identification performance.
3D-CNN learns local geometric features for manufacturability analysis of drilled holes.
problem Capturing distinguishing local features in 3D CAD geometry.
method 3D-CNN with voxel data augmented by surface normals, using 3D gradient-weighted class activation maps.
result Identification of local features critical for manufacturability analysis.
GrateTile optimizes CNN feature map storage for efficient data access.
problem Efficient storage and access of sparse CNN feature maps.
method Divides feature maps into uneven-sized subtensors, compresses and stores them in a compressed yet accessible format.
result Average 55% DRAM bandwidth reduction with minimal indexing overhead.
Improves signature verification accuracy using deep CNN features.
problem Handwritten signature verification accuracy.
method Deep CNN features combined with writer-independent SVM classifier.
result Proposed approach outperforms other WI-HSV methods.
Efficiently extracts local features from whole images using CNNs with pooling layers.
problem Efficiently extracting local features from whole images for various tasks.
method A method to compute patch-based local feature descriptors efficiently in presence of pooling and striding layers for whole images at once, applicable to nearly all existing network architectures.
result Our approach significantly speeds up feature extraction from whole images compared to existing methods.
Boosting CNNs with dynamic feature selection and boosting weights improves accuracy and efficiency.
problem Expensive training of deep CNNs and manual architecture design.
method Dynamic feature selection, importance sampling, and boosting weights integrated into network training.
result Boosted CNN variants consistently outperform conventional CNNs in accuracy and training speed.
Paper uses CNNs to classify heart sounds from short segments.
problem Classifying heart sounds from short segments of individual beats.
method Developed a 1D-CNN and 2D-CNN ensemble for feature learning and score-level fusion.
result ECNN ensemble achieved 89.22% accuracy and 89.94% sensitivity on the PhysioNet CinC 2016 database.
Proposes a new CNN approach for multimodal biometric identification.
problem Improving biometric identification accuracy across multiple modalities.
method Uses a bank of modality-specific CNNs, fuses their outputs, and optimizes the system.
result Significantly outperforms unimodal systems and demonstrates reduction in parameters.
CNNPred uses CNNs to predict stock market movements across multiple markets.
problem Feature extraction and prediction in financial markets.
method CNN-based framework for multiple markets, using various data sources.
result Significant improvement in prediction performance compared to baseline algorithms.
Paper proposes CNN with SIFT for rotation invariant feature extraction.
problem Max-pooling layer discards rotational information, leading to rotation invariance issues.
method Uses SIFT descriptor to capture orientation and spatial relationships.
result Improves feature extraction on MNIST and fashionMNIST datasets.
New adversarial attack method based on deep feature distributions.
problem Adversarial attacks on CNN classifiers using output layer information.
method Modeling and exploiting class-wise and layer-wise deep feature distributions.
result Achieves state-of-the-art transfer-based attack results for undefended ImageNet models.
Dilated CNN improves multivariate time series classification.
problem Multivariate time series classification.
method Transformed multivariate time series into image-like style, applied dilated and strided convolutions.
result Automatic features extracted by dilated CNN are as effective as hand-crafted features.
Proposes a curriculum-based scheme to smooth CNN feature embeddings.
problem Distortion artifacts in early training stages of CNNs.
method Smoothes feature embedding using Gaussian kernels to control high-frequency information.
result Significant performance improvements on various vision tasks.
A method classifies image-sets using convex cones based on CNN features.
problem Image-set classification using CNN features.
method Modeling CNN features as convex cones and measuring geometric similarity.
result Enhanced classification through discriminant space maximization of between-class variance.
This paper establishes a mathematical framework for G-CNNs on homogeneous spaces.
problem Designing equivariant neural networks for data with symmetries.
method Using Mackey's theory on induced representations, the paper presents a general framework for G-CNNs.
result G-CNNs are a universal class of equivariant network architectures.
CNNs encode data augmentation transformations, especially in early layers.
problem Whether neural network features encode data augmentation transformations.
method Systematic approach using pre-trained vision models to predict augmentation transformations.
result Neural network features encode data augmentation transformations, especially in early layers.
Deep CNN classifies EEG-based brain connectivity in schizophrenia.
problem Classifying neuropsychiatric disorders using EEG connectivity.
method Multi-domain connectome CNN framework integrating time and frequency-domain metrics.
result MDC-CNN achieves 93.06% accuracy in schizophrenia classification. New CNN learns depth features from scratch, outperforming transfer learning.
problem Limited depth data for RGB-D scene recognition.
method Bottom-up approach combining local weakly supervised training and global fine-tuning, modified CNN architecture.
result State-of-the-art accuracy on NYU2 and SUN RGB-D datasets.
Paper presents a CNN-RNN method for multi-dimensional emotion recognition in-the-wild.
problem Dimensional emotion recognition in real-world scenarios.
method Pre-training with Aff-Wild and Aff-Wild2, extracting low-, mid-, and high-level features, using RNN subnets in a multi-task framework, and fusion of networks.
result Our approach outperformed state-of-the-art methods using only visual information.
Integrates CNN and GRU for precise stock market risk alerts.
problem Predicting future stock market risks and providing early warnings.
method Uses CNN for feature extraction and GRU for time series analysis.
result Effective early warnings of future stock market risks.
CNN improves spatiotemporal emotion recognition from EEG during music listening.
problem Improving emotion recognition from EEG signals during music listening.
method Conducted a study on CNN and its spatiotemporal feature extraction for emotion recognition.
result CNN outperforms SVM in leave-one-subject-out cross validation.
Compact convolution enhances CNN robustness against adversarial attacks.
problem Improving network robustness against adversarial attacks.
method Learning compact features using L2-Softmax Loss and compact convolution.
result Compact Convolutional Networks (CCNs) neutralize multiple types of adversarial attacks.
LOCO-Reg improves CNN accuracy by promoting feature cohesion near filter centers.
problem Current regularization schemes in CNNs violate the principle that weights near the center of a filter are larger than weights on the outside.
method Introduces Locality-Promoting Regularization (LOCO-Reg) to correct this issue.
result LOCO-Reg yields accuracy gains across multiple architectures and datasets.
New graph CNN layers improve accuracy on graph datasets.
problem Graph data relations are better represented as graphs, not grids.
method Proposed new graph CNN layers for vertex and edge features.
result Improved classification accuracy on graph datasets.
Enhances CNN feature extractors' separation capacity analysis.
problem Understanding the separation capacity of CNNs.
method Extending Cover's function-counting theory, analyzing scattering networks.
result Identifies factors affecting scattering networks' separation capacity.
Paper presents an unsupervised method for object recognition using pretrained CNN and associative memory.
problem Fine-tuning pretrained CNN models for new domains is time-consuming and requires labeled data.
method Uses a pretrained CNN for feature extraction and a Hopfield network associative memory bank for classification.
result Eliminates the need for backpropagation and achieves competitive performance on unseen datasets.
FixyNN splits CNN models into fixed and trainable parts for efficient on-device inference.
problem Energy inefficiency in on-device CNN inference for real-time computer vision.
method Co-designed hardware accelerator platform with transfer learning for training.
result Achieved nearly 2x better energy efficiency than a conventional accelerator.
Early diagnosis, playing an important role in preventing progress and treating the Alzheimer\{'}s disease (AD), is based on classification of features extracted from brain images. The features have to accurately capture main AD-related variations of anatomical brain structures, such as, e.g., ventricles size, hippocamp…
Enhances sound texture in CNN for better acoustic scene classification.
problem Limited understanding of how CNNs perceive audio scenes.
method Used Class Activation Mapping (CAM) to analyze log-Mel features and proposed edge enhancement using DoG and Sobel operators.
result Edge-enhanced log-Mel features improve CNN performance in acoustic scene classification.
CNN-DTW system improves keyword spotting in under-resourced languages.
problem Keyword spotting in nearly zero-resource languages.
method Multilingual bottleneck features, CNN-DTW, DTW template matching, convolutional neural network.
result Multilingual BNFs improve CNN-DTW by 10.9%.
Early diagnosis, playing an important role in preventing progress and treating the Alzheimer's disease (AD), is based on classification of features extracted from brain images. The features have to accurately capture main AD-related variations of anatomical brain structures, such as, e.g., ventricles size, hippocampus …
FixyNN improves energy efficiency of mobile computer vision tasks.
problem High energy consumption of state-of-the-art CNN models on mobile devices.
method Fixed-weight feature extractor and programmable CNN accelerator for transfer learning.
result Achieved up to 26.6 TOPS/W energy efficiency, nearly 2x more efficient than conventional accelerators.
CNNs reveal retinal ganglion cell features, linking visual processing to neuroscience.
problem Understanding what CNNs learn about retinal neuronal circuits.
method Trained CNNs on white noise images to predict neural responses from salamander retinas.
result CNN filters resemble biological retinal components and ganglion cell receptive fields.
JD.com uses a new CNN model to improve ad click prediction.
problem Improving CTR prediction for ads with visual content.
method Proposes Category-specific CNN (CSCNN) to incorporate category knowledge early in the feature extraction process.
result CSCNN outperforms existing methods in CTR prediction.
A new text representation model combines CNN and VAE for better semantic extraction.
problem Difficult to effectively extract semantic features and distinguish polysemy in text data.
method Integrates CNN for feature extraction and VAE for consistent Gaussian distribution.
result The model outperforms traditional classification algorithms in text classification tasks.
New CNNs learn features from unlabeled data for better activity recognition.
problem Limited labeled data for activity recognition leads to poor generalization.
method Semi-supervised CNNs that learn features from raw sensor data.
result Semi-supervised CNNs outperform supervised and traditional methods by up to 18%.
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.
HarDNN detects and protects CNNs from hardware errors.
problem Hardware errors in CNNs can corrupt inference output.
method Statistical error injection and heuristic vulnerability assessment.
result HarDNN improves CNN resilience with minimal additional computation.
The paper analyzes deep neural networks using information theory to improve classification accuracy.
problem Improving classification accuracy in deep neural networks.
method Modeling the output of convolutional filters as a random variable conditioned on class and network structure, computing conditional entropy as a compact code.
result The conditional entropy feature analysis leads to higher classification accuracy than the original CNN.
Method sanitizes IFM in CNN layers to control privacy loss.
problem Controlling privacy loss in CNNs using input feature maps.
method Sample-and-hold approximation scheme to sanitize IFM, unfolding tensors for independence from CNN configuration.
result Control the privacy loss by adjusting the sanitization degree.
In this paper, we predict the likelihood of a player making a shot in basketball from multiagent trajectories. Previous approaches to similar problems center on hand-crafting features to capture domain specific knowledge. Although intuitive, recent work in deep learning has shown this approach is prone to missing impor…
CNN predicts travel modes from GPS trajectories with 84.8% accuracy.
problem Inferring transportation modes from GPS data.
method CNN architecture for automatic feature extraction from raw GPS data.
result Highest accuracy of 84.8% achieved through ensemble of CNN configurations.
The paper analyzes how CNNs interpret NLP tasks and identify linguistic features.
problem Understanding how CNNs capture linguistic features in NLP tasks.
method Visualization techniques and error analysis to interpret CNNs.
result Identified how CNNs capture different linguistic features and their impact on model performance.
This paper improves land cover classification using global spatial features in CNN.
problem Limited classification accuracy and universality of traditional remote sensing image classification methods.
method Integrates global spatial features into a dual-branch CNN for hyperspectral/SAR imagery classification.
result The proposed method outperforms traditional single-channel CNN methods.