Results 101 to 110 of about 36,322 (258)

On-the-fly Network Pruning for Object Detection

CoRR, 2016
Object detection with deep neural networks is often performed by passing a few thousand candidate bounding boxes through a deep neural network for each image. These bounding boxes are highly correlated since they originate from the same image. In this paper we investigate how to exploit feature occurrence at the image scale to prune the neural network ...
Marc Masana, Joost van de Weijer 0001, Andrew D. Bagdanov   +2 more
openaire   +2 more sources

Overcoming the Phase Discontinuity Problem in RIS Optimization: A Robust and Scalable Phase‐Aware Deep Regression Framework via Hybrid CNN–LSTM Architecture

Advanced Intelligent Systems, EarlyView.
The phase discontinuity problem—where the cyclic nature of phase angles causes catastrophic errors near the ±π boundary—is a fundamental obstacle in learning‐based reconfigurable intelligent surface (RIS) optimization. A phase‐aware hybrid CNN–LSTM framework resolves this by decomposing phase predictions into sine–cosine components, mapping circular ...
Seda Savaşçı Şen, Ali Çalhan, Murtaza Cicioğlu, Oğuzhan Demiryürek   +3 more
wiley   +1 more source

PrONe-RS: A Graph-Isomorphic Pruning With Pareto-Driven Optimization Framework for Network Sizing in Remote Sensing LULC Applications

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Deep learning (DL) models have demonstrated remarkable performance in remote sensing (RS) land use land cover (LULC) classification. Yet, their high computational complexity demands can limit their deployment in resource-constrained edge computing ...
Nikos Temenos, Anastasios Temenos, Charalampos Zafeiropoulos, Dimitris Kalogeras, Anastasios Doulamis, Nikolaos Doulamis, Ioannis Rallis, Manthos Bimpas   +7 more
doaj   +1 more source

A Novel Dynamic Network Pruning via Smooth Initialization and its Potential Applications in Machine Learning Based Security Solutions

IEEE Access, 2019
With the development of artificial intelligence technology, the demand for new digital security and privacy solutions is growing importantly. Inspired by the synaptic pruning in mammalian brains, we develop a network pruning method, called dynamic ...
Lei Wu, Han Yue, Peng Chen, De-An Wu, Qiu Jin   +4 more
doaj   +1 more source

Co‐Design of Stretchable Fabric Sensors and Tiny Machine Learning for Human Interface Device‐Based Edge‐Intelligent Wearable Gloves

Advanced Intelligent Systems, EarlyView.
A flexible smart glove integrated with textile sensors enables real‐time hand gesture recognition for human–machine interaction. Using edge AI inference and Bluetooth communication, the system translates finger movements into human interface device commands to control endpoint devices efficiently.
Chi Cuong Vu, Tuan Nghia Nguyen, Minh‐Thanh Le   +2 more
wiley   +1 more source

Application of discrete Ricci curvature in pruning randomly wired neural networks: a case study with chest x-ray classification of COVID-19

Journal of Physics: Complexity
Randomly wired neural networks (RWNNs) serve as a valuable testbed for investigating the impact of network topology in deep learning by capturing how different connectivity patterns impact both learning efficiency and model performance. At the same time,
Pavithra Elumalai, Sudharsan Vijayaraghavan, Madhumita Mondal, Areejit Samal   +3 more
doaj   +1 more source

Tailored Channel Pruning: Achieve Targeted Model Complexity Through Adaptive Sparsity Regularization

IEEE Access
In deep learning, the size and complexity of neural networks have been rapidly increased to achieve higher performance. However, this poses a challenge when utilized in resource-limited environments, such as mobile devices, particularly when trying to ...
Suwoong Lee, Yunho Jeon, Seungjae Lee, Junmo Kim   +3 more
doaj   +1 more source

High-Efficient Parameter-Pruning Algorithm of Decision Tree for Large Dataset [PDF]

Jisuanji gongcheng
Decision tree(DT) have a good effect on data classification but easily develop overfitting. The solution to this problem is to prune the DT. However, the pruning algorithm has shortcomings; for example, prepruning is prone to underfitting, the ...
Zhaoxian XIE, Xingmin ZOU, Wenjing ZHANG
doaj   +1 more source

Vernacular Futurism: How Persian Language Users Imagine AI

AI &Innovation, EarlyView.
ABSTRACT Public discourse about artificial intelligence increasingly unfolds through compressed forecasts, moral warnings, and everyday speculation circulating at platform speed. This study examines how Persian language users on X construct and contest AI futures, analyzing a corpus of 4741 posts collected between January 2023 and December 2025, with ...
Arthur Asa Berger, Ehsan Shahghasemi
wiley   +1 more source

Playing in the Dark: Invisible Chess as a Laboratory for Strategic AI

AI &Innovation, EarlyView.
This paper shows that strategic AI evaluated on perfect‐information benchmarks can be brittle in real adversarial settings. By using invisible chess as a benchmark for hidden state and deception, it argues for stricter testing, human oversight, and more cautious governance of high‐stakes AI systems.
Paolo Ciancarini
wiley   +1 more source