convolutional neural network
PulseAugur coverage of convolutional neural network — every cluster mentioning convolutional neural network across labs, papers, and developer communities, ranked by signal.
6 天有情绪数据
-
New method learns stability landscapes from network topology
Researchers have introduced a new method for analyzing synchronization networks by learning "stability landscapes" directly from graph topology. This approach uses a graph-to-image prediction paradigm, where a Graph Neu…
-
Smartwatches detect drunk driving using motion and heart rate data
Researchers have developed a system using off-the-shelf smartwatches to detect alcohol-impaired driving. The system analyzes wrist accelerometer data and heart rate variability to identify intoxication levels. In a test…
-
New active learning methods boost data efficiency for deep learning
Researchers have developed four new hybrid sampling methods for active learning in deep learning models, aiming to improve efficiency in data labeling for computer vision tasks. These methods combine the selection of bo…
-
Algebraic ML framework matches CNNs, XGBoost on small datasets
Researchers have developed a new framework called Algebraic Machine Learning (AML) that learns through algebraic structure decomposition, bypassing traditional numerical optimization. In evaluations, AML demonstrated co…
-
Deep learning models show promise for analyzing retinal images
Researchers have explored the use of deep learning models, including convolutional neural networks, vision transformers, and foundation models, for analyzing ultra-widefield (UWF) retinal images. The study focused on th…
-
CNNs achieve 96% accuracy classifying partial discharge using novel AWA patterns
Researchers have developed a novel Amplitude-Width-Area (AWA) pattern representation to analyze partial discharge (PD) pulses under switching-voltage excitation. This method maps PD pulses into visual patterns using amp…
-
Vision Transformers and CNNs Compared for Land Use Classification
A new research paper compares the effectiveness of Vision Transformers (ViTs) and Convolutional Neural Networks (CNNs) for land use scene classification using remote sensing imagery. The study evaluated AlexNet and ViT …
-
New method tackles CNNs' reliance on spurious correlations
Researchers have developed Deep Attention Reweighting (DAR), a novel post-hoc method to improve the generalization and fairness of Convolutional Neural Networks (CNNs). DAR addresses the issue of CNNs exploiting spuriou…
-
New AI models for Arabic handwriting recognition vulnerable to attacks
Researchers have developed lightweight embedded ConvNet ensembles to improve Arabic handwritten character recognition, achieving accuracy comparable to larger models. A separate study investigated the security of these …
-
New system fuses ToF and IR sensors for efficient gesture recognition
Researchers have developed a novel gesture recognition system designed for resource-constrained devices like smart eyewear. This system efficiently fuses data from low-resolution Time-of-Flight and Infrared thermal sens…
-
CNN framework tests General Relativity using gravitational wave data
Researchers have developed a convolutional neural network (CNN) framework to test General Relativity using gravitational wave data. By training the CNN on simulated beyond-GR waveforms, they found that using a response …
-
Deep learning receiver boosts asynchronous comms in control networks
Researchers have developed a novel deep learning-based receiver designed to improve asynchronous grant-free random access in control-to-control communication networks. This system utilizes a convolutional neural network…
-
Foundation model learns from Dutch satellite data for global benchmarks
Researchers have developed a new foundation model for high-resolution remote sensing data, specifically trained on satellite images of the Netherlands. This model combines Convolutional Neural Networks and Vision Transf…
-
New OUIDecay method adapts CNN regularization layer-by-layer
Researchers have introduced OUIDecay, a novel adaptive weight decay method for convolutional neural networks. This technique dynamically adjusts regularization strength for each layer based on online activation patterns…
-
New vehicle classifier combines spatial awareness with explainability
Researchers have developed an enhanced vehicle classification system that incorporates spatial awareness of vehicle parts. This new method builds upon a previous approach by constructing spatial probability maps for eac…
-
CNN achieves 91.79% accuracy for Hindi keyword spotting in speech recognition
Researchers have developed a keyword spotting system for Hindi speech recognition using a Convolutional Neural Network (CNN). The system was trained on 40,000 audio samples and utilizes Mel Frequency Cepstral Coefficien…
-
Deep learning model predicts full-chip CMP nanotopography with nanometer accuracy
Researchers have developed a novel deep learning model to predict the full-chip post-Chemical-Mechanical Polishing (CMP) nanotopography with nanometer-scale accuracy. This model combines data from White Light Interferom…
-
2D convolutions speed up EEG signal classification
Researchers have explored using 2D spatiotemporal convolutions for classifying EEG signals, an alternative to the common practice of concatenating 1D spatial and temporal convolutions. Their findings indicate that 2D co…
-
New BerLU activation function improves deep learning stability and efficiency
Researchers have introduced a new activation function called the Bernstein Linear Unit (BerLU) that aims to improve the stability and efficiency of deep neural networks. By utilizing Bernstein polynomials, BerLU creates…
-
Researchers introduce PREMAP2 for efficient neural network certification
Researchers have developed PREMAP2, an enhanced algorithm for approximating neural network preimages, significantly improving scalability and efficiency. This new method extends the capabilities of its predecessor, PREM…