deep neural network
PulseAugur coverage of deep neural network — every cluster mentioning deep neural network across labs, papers, and developer communities, ranked by signal.
8 day(s) with sentiment data
-
New augmentation method improves spatial transcriptomics imputation
Researchers have developed SNR-ST-Mix, a novel data augmentation framework for spatial transcriptomics imputation using deep neural networks. This method addresses limitations in current augmentation strategies by ensur…
-
Terastal framework optimizes DNN scheduling on heterogeneous accelerators
Researchers have developed Terastal, a new framework designed to improve the scheduling of multiple deep neural networks (DNNs) on heterogeneous accelerators for soft real-time applications. The system addresses latency…
-
Deep Gaussian Processes show non-Gaussian limits below critical threshold
Researchers have identified a critical threshold in compositional Gaussian Processes (GPs) that determines whether their behavior in deep models becomes degenerate or non-trivial. The study establishes a sharp bandwidth…
-
New framework unifies singular learning theory and information geometry
Researchers have developed a new framework called Geometric Singular Learning that bridges singular learning theory and information geometry. This approach introduces the concept of a "dead direction" to unify parameter…
-
Deep learning aids radiomic feature selection for lung cancer detection
Researchers have developed a new framework called Gradient-Loss Recursive Feature Elimination (GL-RFE) to improve the selection of radiomic features for lung cancer stage detection. This method uses a deep neural networ…
-
Physical adversarial patches fool aerial vehicle detectors
Researchers have developed a method to create physical adversarial patches that can fool deep neural network-based aerial vehicle detectors. These patches are optimized digitally with constraints for printability and sm…
-
New CAFD method uses VLMs for efficient DNN fault detection
Researchers have developed a new method called Concept-Aware Fault Detection (CAFD) to identify errors in Deep Neural Networks (DNNs). CAFD integrates various data sources, including a novel "Concept Failure Ratio" deri…
-
New AI model uses WTA bottlenecks for symbolic representation
Researchers have developed a novel deep learning model that utilizes Winner-Take-All (WTA) bottlenecks to enforce the extraction of disentangled symbolic representations in multi-task learning. This approach, inspired b…
-
Machine learning models improve patient mortality prediction using medical notes
Researchers have developed a new Deep Neural Network (DNN) model with a pooling mechanism to improve the prediction of patient mortality after hospital discharge. This model leverages unstructured medical notes, which o…
-
AI models adapt to detect synthetic fingerprints with few-shot learning
Researchers have developed a new method for detecting synthetic fingerprints generated by artificial intelligence, addressing the increasing realism of these fakes. The approach treats synthetic fingerprint detection as…
-
LUNA architecture accelerates quantum qubit readout with LUT-based neural networks
Researchers have developed LUNA, a novel neural architecture designed for faster and more cost-effective qubit readout in quantum computing. This system integrates low-cost integrator-based preprocessing with Look-Up Ta…
-
Tessera offers secure, near-line-rate weight streaming for edge AI accelerators
Researchers have developed Tessera, a new architecture designed to securely stream model weights to edge accelerators in Unified Memory Architecture (UMA) systems. This approach addresses the challenge of protecting pro…
-
New deep neural network framework offers interpretable survival data analysis
Researchers have introduced FLEXI-Haz, a novel deep neural network framework designed for survival data analysis with a partially linear regression structure. This method distinguishes itself by maintaining interpretabi…
-
New HGQ-LUT and da4ml methods speed up DNN training and FPGA deployment
Researchers have developed HGQ-LUT, a new method for training lookup-table (LUT) based neural networks that significantly speeds up the training process, making it over 100 times faster on modern GPUs. This approach int…
-
New theory shows compact datasets can be made linearly separable by DNNs
Researchers have developed a theory for relocating compact sets in $\mathbb{R}^n$ to arbitrary target domains using diffeomorphisms. This work demonstrates that such collections can be embedded into $\mathbb{R}^{n+1}$ t…