Results 101 to 110 of about 1,314,055 (250)

A Multi-Class ECG Signal Classifier Using a Binarized Depthwise Separable CNN with the Merged Convolution–Pooling Method

Sensors
Binarized convolutional neural networks (bCNNs) are favored for the design of low-storage, low-power cardiac arrhythmia classifiers owing to their high weight compression rate.
Rui Zhang, Ranran Zhou, Zuting Zhong, Haifeng Qi, Yong Wang   +4 more
doaj   +1 more source

In Situ Quantization with Memory‐Transistor Transfer Unit Based on Electrochemical Random‐Access Memory for Edge Applications

Advanced Science, EarlyView.
By combining ionic nonvolatile memories and transistors, this work proposes a compact synaptic unit to enable low‐precision neural network training. The design supports in situ weight quantization without extra programming and achieves accuracy comparable to ideal methods. This work obtains energy consumption advantage of 25.51× (ECRAM) and 4.84× (RRAM)
Zhen Yang, Yuxiang Yang, Baiqian Wang, Yaoyu Tao, Zelun Pan, Lei Cai, Teng Zhang, Longhao Yan, Xianbin Li, Yuchao Yang   +9 more
wiley   +1 more source

Deep Learning Prediction of Surface Roughness in Multi‐Stage Microneedle Fabrication: A Long Short‐Term Memory‐Recurrent Neural Network Approach

Advanced Intelligent Discovery, EarlyView.
A sequential deep learning framework is developed to model surface roughness progression in multi‐stage microneedle fabrication. Using real‐world experimental data from 3D printing, molding, and casting stages, an long short‐term memory‐based recurrent neural network captures the cumulative influence of geometric parameters and intermediate outputs ...
Abdollah Ahmadpour, Sara Soltanabadi Farshi, Tuanna Ozcan, Amir Mohammad Ketabchi, Alper Kiraz, Savas Tasoglu   +5 more
wiley   +1 more source

Combinatorial Attacks on Binarized Neural Networks

, 2018
Binarized Neural Networks (BNNs) have recently attracted significant interest due to their computational efficiency. Concurrently, it has been shown that neural networks may be overly sensitive to "attacks" - tiny adversarial changes in the input - which may be detrimental to their use in safety-critical domains.
Khalil, Elias B., Gupta, Amrita, Dilkina, Bistra   +2 more
openaire   +2 more sources

Scaling Binarized Neural Networks on Reconfigurable Logic [PDF]

Proceedings of the 8th Workshop and 6th Workshop on Parallel Programming and Run-Time Management Techniques for Many-core Architectures and Design Tools and Architectures for Multicore Embedded Computing Platforms, 2017
Binarized neural networks (BNNs) are gaining interest in the deep learning community due to their significantly lower computational and memory cost. They are particularly well suited to reconfigurable logic devices, which contain an abundance of fine-grained compute resources and can result in smaller, lower power implementations, or conversely in ...
Fraser, Nicholas J., Umuroglu, Yaman, Gambardella, Giulio, Blott, Michaela, Leong, Philip W., Vissers, Kees, Jahre, Magnus   +6 more
openaire   +3 more sources

CrossMatAgent: AI‐Assisted Design of Manufacturable Metamaterial Patterns via Multi‐Agent Generative Framework

Advanced Intelligent Discovery, EarlyView.
CrossMatAgent is a multi‐agent framework that combines large language models and diffusion‐based generative AI to automate metamaterial design. By coordinating task‐specific agents—such as describer, architect, and builder—it transforms user‐provided image prompts into high‐fidelity, printable lattice patterns.
Jie Tian, Martin Taylor Sobczak, Dhanush Patil, Jixin Hou, Lin Pang, Arunachalam Ramanathan, Libin Yang, Xianyan Chen, Yuval Golan, Xiaoming Zhai, Hongyue Sun, Kenan Song, Xianqiao Wang   +12 more
wiley   +1 more source

In‐Memory Binary Vector–Matrix Multiplication Based on Complementary Resistive Switches

Advanced Intelligent Systems, 2020
This work studies a computation in‐memory concept for binary multiply‐accumulate operations based on complementary resistive switches (CRS). By exploiting the in‐memory boolean exclusive OR (XOR) operation of single CRS devices, the Hamming Distance (HD)
Tobias Ziegler, Rainer Waser, Dirk J. Wouters, Stephan Menzel   +3 more
doaj   +1 more source

Harnessing Phase Dynamics Across Diverse Frequencies with Multifrequency Oscillatory Neural Networks

Advanced Intelligent Discovery, EarlyView.
Oscillatory Neural Networks (ONNs) are an emerging computing paradigm that encodes information in the phases of coupled oscillators. Traditionally, ONNs have been investigated using homogeneous frequency oscillators. However, physical hardware implementations are inherently subject to frequency mismatches, device variability, and nonuniformities.
Nil Dinç, Filip Sabo, Aida Todri‐Sanial   +2 more
wiley   +1 more source

An Intrusion Detection System for 5G SDN Network Utilizing Binarized Deep Spiking Capsule Fire Hawk Neural Networks and Blockchain Technology

Future Internet
The advent of 5G heralds unprecedented connectivity with high throughput and low latency for network users. Software-defined networking (SDN) plays a significant role in fulfilling these requirements. However, it poses substantial security challenges due
Nanavath Kiran Singh Nayak, Budhaditya Bhattacharyya   +1 more
doaj   +1 more source

Advancing Efficient Error Reduction in DNA Data Storage Systems with Deep Learning‐Based Denoising Models

Advanced Intelligent Discovery, EarlyView.
Deep learning‐based denoising models are applied to DNA data storage systems to enhance error reduction and data fidelity. By integrating DnCNN with DNA sequence encoding methods, the study demonstrates significant improvements in image quality and correction of substitution errors, revealing a promising path toward robust and efficient DNA‐based ...
Seongjun Seo, Anshula Tandon, Nguyen Kim Uyen, Quoc Nam Le, Jee‐Hyong Lee, Sung Ha Park   +5 more
wiley   +1 more source