Results 71 to 80 of about 7,952,537 (299)

Control flow graph based code optimization using graph neural networks

open access: yesFrontiers in Robotics and AI
Selecting a good set of optimization flags requires extensive effort and expert input. While most of the prior research considers using static, spatial, or dynamic features, some of the latest research directly applied deep neural networks to source code.
Melih Peker, Ozcan Ozturk
doaj   +1 more source

Dynamic Correlation Adjacency-Matrix-Based Graph Neural Networks for Traffic Flow Prediction

open access: yesSensors, 2023
Modeling complex spatial and temporal dependencies in multivariate time series data is crucial for traffic forecasting. Graph convolutional networks have proved to be effective in predicting multivariate time series. Although a predefined graph structure
Junhua Gu   +4 more
doaj   +1 more source

All‐in‐One Analog AI Hardware: On‐Chip Training and Inference with Conductive‐Metal‐Oxide/HfOx ReRAM Devices

open access: yesAdvanced Functional Materials, EarlyView.
An all‐in‐one analog AI accelerator is presented, enabling on‐chip training, weight retention, and long‐term inference acceleration. It leverages a BEOL‐integrated CMO/HfOx ReRAM array with low‐voltage operation (<1.5 V), multi‐bit capability over 32 states, low programming noise (10 nS), and near‐ideal weight transfer.
Donato Francesco Falcone   +11 more
wiley   +1 more source

Biodegradable and Biocompatible Functional Polymers for Biomedical Applications

open access: yesAdvanced Functional Materials, EarlyView.
Biodegradable and biocompatible functional polymers integrate electrical, mechanical, and stimuli‐responsive functionalities while enabling programmed degradation under physiological conditions. This review introduces recent advances in conductive, shape‐memory, self‐healing, photocurable, and adhesive polymer systems, emphasizing material design ...
Won Bae Han   +5 more
wiley   +1 more source

Dynamic Heterogeneous Graph Learning: An Adaptive Research Academic Network

open access: yesIEEE Access
Accurately assessing the impact of academic achievements holds significant importance for scholars in their literature review process and for the retrieval and recommendation of scientific research databases.
Minghong Yang, Zhenhua Yang
doaj   +1 more source

Multimode Oxide‐Based Optoelectronic Memtransistor for In‐Sensor Vision Processing

open access: yesAdvanced Functional Materials, EarlyView.
A multimode optoelectronic memtransistor (OEMT) is demonstrated for vision explainable artificial intelligence (VXAI) hardware. By integrating optical sensing, electrical masking, and non‐volatile memory, the device enables key operations required for generating saliency information.
Min Gu Lee   +10 more
wiley   +1 more source

Memristive‐Gated RC‐Delay Synaptic Transistors for Time‐Encoded Analog in‐Memory Computing

open access: yesAdvanced Functional Materials, EarlyView.
A Memristive‐Gated Transistor for Time‐Encoded Analog In‐Memory Computing — By exploiting the RC delay of a self‐rectifying interface‐type memristor, nonlinear I–V distortion is structurally bypassed, enabling 3‐bit nonvolatile memory, spike‐timing‐based analog encoding, and hardware‐calibrated reservoir‐computing validation within a unified device ...
Yun‐Seo Shin   +7 more
wiley   +1 more source

Path integral based convolution and pooling for graph neural networks [PDF]

open access: yes, 2020
Graph neural networks (GNNs) extends the functionality of traditional neural networks to graph-structured data. Similar to CNNs, an optimized design of graph convolution and pooling is key to success.
Lio P.   +4 more
core  

Graph-generative neural network for EEG-based epileptic seizure detection via discovery of dynamic brain functional connectivity

open access: yesScientific Reports, 2022
Dynamic complexity in brain functional connectivity has hindered the effective use of signal processing or machine learning methods to diagnose neurological disorders such as epilepsy. This paper proposed a new graph-generative neural network (GGN) model
Zhengdao Li   +4 more
doaj   +1 more source

Analog Classification of Human Hand Gestures Exploiting Dendritic Integration With an Organic Artificial Soma

open access: yesAdvanced Functional Materials, EarlyView.
An organic neuromorphic architecture for the classification of human motor behavior is presented and validated. It performs somatic integration by linearly combining the activity from three muscles. An investigation of synaptic weights is presented and discussed in relationship with the classification performance.
Ilenia Sergi   +7 more
wiley   +1 more source

Home - About - Disclaimer - Privacy