Suchergebnisse - "Memory efficient"

Deep Pruner and Adaptive Cost Volume Multiview Stereo Network for 3D Reconstruction

Autoren: Junaid Jamshid Wan Wanggen Khurram Shahzad et al.

Quelle: IEEE Access, Vol 13, Pp 28777-28788 (2025)

Schlagwörter: Aggregated cost volume, pruning, feature network, 3D reconstruction, Electrical engineering. Electronics. Nuclear engineering, memory efficient, TK1-9971

Zugangs-URL: https://doaj.org/article/57ed44961a76421e9947f3e2908b1732

Contrastive Feature Bin Loss for Monocular Depth Estimation

Autoren: Jihun Song Yoonsuk Hyun

Quelle: IEEE Access, Vol 13, Pp 49584-49596 (2025)

Schlagwörter: memory efficient training, contrastive learning, Electrical engineering. Electronics. Nuclear engineering, Monocular depth estimation, TK1-9971

Zugangs-URL: https://doaj.org/article/04d13fcf9e9b43eaaac90bf937e3c9dd

Deep Time Series Intelligent Framework for Power Data Asset Evaluation

Autoren: Lihong Ge Xin Li Li Wang et al.

Quelle: IEEE Access, Vol 13, Pp 86277-86289 (2025)

Schlagwörter: sophisticated convolutional neural network, Power data assets, time-series convolutional memory efficient network, long-and short-term time-series network, Electrical engineering. Electronics. Nuclear engineering, expressway network, TK1-9971

Zugangs-URL: https://doaj.org/article/99c5355c7e0b4e7bb5c00d8f298ccf51

Folded Context Condensation in Path Integral Formalism for Infinite Context Transformers

Autoren: Won-Gi Paeng Daesuk Kwon Kyungwon Jeong et al.

Quelle: IEEE Access, Vol 13, Pp 77582-77592 (2025)

Schlagwörter: FOS: Computer and information sciences, feed-forward network (FNN), Computer Science - Machine Learning, Computer Science - Computation and Language, Computer Science - Artificial Intelligence, quantum mechanics, Attention mechanism, Computer Science - Neural and Evolutionary Computing, FOS: Physical sciences, TK1-9971, Machine Learning (cs.LG), condensation, High Energy Physics - Phenomenology, High Energy Physics - Phenomenology (hep-ph), Artificial Intelligence (cs.AI), Electrical engineering. Electronics. Nuclear engineering, Neural and Evolutionary Computing (cs.NE), memory-efficient, Computation and Language (cs.CL), path integral

Zugangs-URL: http://arxiv.org/abs/2405.04620
https://doaj.org/article/0a0d5063677b425ebeb8c5aa544172e9

NeuroFlux: Memory-Efficient CNN Training Using Adaptive Local Learning

Autoren: Dhananjay Saikumar Blesson Varghese University of St Andrews.School of Computer Science

Weitere Verfasser: Dhananjay Saikumar Blesson Varghese University of St Andrews.School of Computer Science

Quelle: Proceedings of the Nineteenth European Conference on Computer Systems. :999-1015

Schlagwörter: QA75, Computer Science - Machine Learning, QA75 Electronic computers. Computer science, Memory efficient training, Edge computing, NIS, CNN training, Local learning, 3rd-NDAS

Dateibeschreibung: application/pdf

Zugangs-URL: http://arxiv.org/abs/2402.14139
https://hdl.handle.net/10023/29805

Symbiotic Tuning: A Simple Approach for Enhancing Task Performance of Side-Tuning

Autoren: Zhi-Quan Feng Ying-Jia Lin Kun-Ta Chuang et al.

Quelle: IEEE Access, Vol 13, Pp 198471-198481 (2025)

Schlagwörter: Deep learning, transformer model, natural language processing, parameter-efficient fine-tuning, memory-efficient fine-tuning, natural languege understanding, Electrical engineering. Electronics. Nuclear engineering, TK1-9971

Dateibeschreibung: electronic resource

Relation: https://ieeexplore.ieee.org/document/11227100/; https://doaj.org/toc/2169-3536

Zugangs-URL: https://doaj.org/article/eb02d74b0763457991d3678c8f0e0a38

Edge-Optimized CNNs: A Co-Designed Software-Hardware Framework for Lightweight Deep Learning

Autoren: Zahra Waheed Yazeed Alkhrijah Shahzad Khalid et al.

Quelle: IEEE Access, Vol 13, Pp 184679-184693 (2025)

Schlagwörter: Software/hardware co-design, memory-efficient network compression, parallel hardware architecture, low resource-utilization, Electrical engineering. Electronics. Nuclear engineering, TK1-9971

Dateibeschreibung: electronic resource

Relation: https://ieeexplore.ieee.org/document/11214356/; https://doaj.org/toc/2169-3536

Zugangs-URL: https://doaj.org/article/dd7413238b4742f2ab651b4ccfce1cc3

FracGrad: A Discretized Riemann–Liouville Fractional Integral Approach to Gradient Accumulation for Deep Learning

Autoren: Minhyeok Lee

Quelle: Fractal and Fractional, Vol 9, Iss 11, p 733 (2025)

Schlagwörter: gradient accumulation, fractional calculus, memory-efficient training, stochastic optimization, power-law weighting, deep learning, Thermodynamics, QC310.15-319, Mathematics, QA1-939, Analysis, QA299.6-433

Dateibeschreibung: electronic resource

Relation: https://www.mdpi.com/2504-3110/9/11/733; https://doaj.org/toc/2504-3110

Zugangs-URL: https://doaj.org/article/529f556212594d0d90cabdac8df54c89

The ASC Module: A GPU Memory-Efficient, Physiology-Aware Approach for Improving Segmentation Accuracy on Poorly Contrast-Enhanced CT Scans—A Preliminary Study

Autoren: Zuoyuan Zhao Toru Higaki Yanlei Gu et al.

Quelle: Bioengineering, Vol 12, Iss 9, p 974 (2025)

Schlagwörter: GPU memory-efficient, image segmentation, CT images, poor/non-contrasted CT, spatial structure, deep learning, Technology, Biology (General), QH301-705.5

Dateibeschreibung: electronic resource

Relation: https://www.mdpi.com/2306-5354/12/9/974; https://doaj.org/toc/2306-5354

Zugangs-URL: https://doaj.org/article/7d0790676cde43d6bce1c9e0abd3bbee

Overview of Memory-Efficient Architectures for Deep Learning in Real-Time Systems

Autoren: Bilgin Demir Ervin Domazet Daniela Mechkaroska

Quelle: Engineering Proceedings, Vol 104, Iss 1, p 77 (2025)

Schlagwörter: deep learning, memory-efficient architectures, real-time data processing, low-power devices, quantization, parameter pruning, Engineering machinery, tools, and implements, TA213-215

Dateibeschreibung: electronic resource

Relation: https://www.mdpi.com/2673-4591/104/1/77; https://doaj.org/toc/2673-4591

Zugangs-URL: https://doaj.org/article/a423bb8a918f45d5a5b853edb05cb209

Entropy-Guided KV Caching for Efficient LLM Inference

Autoren: Heekyum Kim Yuchul Jung

Quelle: Mathematics, Vol 13, Iss 15, p 2366 (2025)

Schlagwörter: LLM, KV cache, transformer, LLM inference optimization, attention entropy, memory-efficient caching, Mathematics, QA1-939

Dateibeschreibung: electronic resource

Relation: https://www.mdpi.com/2227-7390/13/15/2366; https://doaj.org/toc/2227-7390

Zugangs-URL: https://doaj.org/article/8b5f893e0d4642be98b40e1672c8e4d7

Physical data embedding for memory efficient AI

Autoren: Callen MacPhee Yiming Zhou Bahram Jalali

Quelle: Machine Learning: Science and Technology, Vol 6, Iss 4, p 045018 (2025)

Schlagwörter: physics-AI symbiosis, interpretable AI, physics-inspired algorithms, physics-based neural networks, memory-efficient AI, Computer engineering. Computer hardware, TK7885-7895, Electronic computers. Computer science, QA75.5-76.95

Relation: https://doi.org/10.1088/2632-2153/ae0f37; https://doaj.org/toc/2632-2153; https://doaj.org/article/b97d56612e9342afadd949cf144179ee

Verfügbarkeit: https://doi.org/10.1088/2632-2153/ae0f37
https://doaj.org/article/b97d56612e9342afadd949cf144179ee

A Reinduction-Based Approach for Efficient High Utility Itemset Mining from Incremental Datasets

Autoren: Pushp Sra Satish Chand

Quelle: Data Science and Engineering, Vol 9, Iss 1, Pp 73-87 (2023)

Schlagwörter: Knowledge discovery, Incremental datasets, Memory efficient mining, Electronic computers. Computer science, 0202 electrical engineering, electronic engineering, information engineering, Information technology, QA75.5-76.95, 02 engineering and technology, T58.5-58.64, High utility itemset mining, Data mining

Zugangs-URL: https://doaj.org/article/311f5866752845f78a2fcbbc5151eeb8

Automatic recognition of white blood cell images with memory efficient superpixel metric GNN: SMGNN

Autoren: Yuanhong Jiang Yiqing Shen Yuguang Wang et al.

Quelle: Mathematical Biosciences and Engineering, Vol 21, Iss 2, Pp 2163-2188 (2024)

Schlagwörter: Cytoplasm, 0211 other engineering and technologies, 02 engineering and technology, memory efficient model, white blood cell segmentation, cell type classification, 3. Good health, QA1-939, Image Processing, Computer-Assisted, Leukocytes, 0202 electrical engineering, electronic engineering, information engineering, Neural Networks, Computer, gnn, superpixel metric learning, TP248.13-248.65, Mathematics, Algorithms, Biotechnology

Zugangs-URL: https://pubmed.ncbi.nlm.nih.gov/38454678
https://doaj.org/article/6bbb842d0c624be59ef67ad20c5902f4

p-Meta: Towards On-device Deep Model Adaptation

Autoren: QU, Zhongnan ZHOU, Zimu TONG, Yongxin et al.

Quelle: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. :1441-1451

Schlagwörter: FOS: Computer and information sciences, Computer Science - Machine Learning, deep neural networks, Computer Vision and Pattern Recognition (cs.CV), meta learning, Computer Science - Computer Vision and Pattern Recognition, 0202 electrical engineering, electronic engineering, information engineering, memory-efficient training, 02 engineering and technology, Numerical Analysis and Scientific Computing, Machine Learning (cs.LG)

Dateibeschreibung: application/pdf

Zugangs-URL: http://arxiv.org/abs/2206.12705

Fully invertible hyperbolic neural networks for segmenting large-scale surface and sub-surface data

Autoren: Bas Peters Eldad Haber Keegan Lensink

Quelle: Artificial Intelligence in Geosciences, Vol 5, Iss , Pp 100087- (2024)

Schlagwörter: Invertible neural networks, Large scale deep learning, Memory efficient deep learning, Geography (General), G1-922, Information technology, T58.5-58.64

Dateibeschreibung: electronic resource

Relation: http://www.sciencedirect.com/science/article/pii/S2666544124000285; https://doaj.org/toc/2666-5441

Zugangs-URL: https://doaj.org/article/8c33c70e82444bcd9431b1e5e92fc40c

Spiral Doorway Prime Generation: A Deterministic Framework for the Emergence of Prime Structure Verified Generation of 100 Million Primes at 100% Accuracy via Geometric Shell Recursion

Autoren: Needham, Eric

Schlagwörter: Prime Generation, φ-Harmonic Geometry, Deterministic Algorithms, Spiral Doorway Method, Bootstrap Calibration, Recursive Vacuum Theory, Fibonacci Resonance, Shell Structure, Prime Number Theorem, Computational Number Theory, Predictive Framework, Primality Verification, Memory-Efficient Algorithms, Prime Gap Structure, Emergent Number Fields, Eric Needham

Relation: https://zenodo.org/records/16656771; oai:zenodo.org:16656771; https://doi.org/10.5281/zenodo.16656771

Verfügbarkeit: https://doi.org/10.5281/zenodo.16656771
https://zenodo.org/records/16656771

Resource-Restricted Environments Based Memory-Efficient Compressed Convolutional Neural Network Model for Image-Level Object Classification

Autoren: Zahra Waheed Shehzad Khalid Syed Mursleen Riaz et al.

Quelle: IEEE Access, Vol 11, Pp 1386-1406 (2023)

Schlagwörter: pruning, Memory-efficient network compression, 0202 electrical engineering, electronic engineering, information engineering, quantization, Electrical engineering. Electronics. Nuclear engineering, 02 engineering and technology, image-level object classification, resource-restricted edge-devices, TK1-9971

Zugangs-URL: https://doaj.org/article/d23a448e391f448087bff72e78a5ecdf

Toward Memory-Efficient and Interpretable Factorization Machines via Data and Model Binarization

Autoren: Yu Geng Liang Lan William K. Cheung

Quelle: IEEE Access, Vol 11, Pp 128633-128643 (2023)

Schlagwörter: memory-efficient design, factorization machines, Electrical engineering. Electronics. Nuclear engineering, Binarization, interpretability, TK1-9971

Zugangs-URL: https://doaj.org/article/1f59538ca1fd4f67bdb44480f5d86141

Automatic recognition of white blood cell images with memory efficient superpixel metric GNN: SMGNN

Autoren: Yuanhong Jiang Yiqing Shen Yuguang Wang et al.

Quelle: Mathematical Biosciences and Engineering, Vol 21, Iss 2, Pp 2163-2188 (2024)

Schlagwörter: gnn, superpixel metric learning, memory efficient model, white blood cell segmentation, cell type classification, Biotechnology, TP248.13-248.65, Mathematics, QA1-939

Dateibeschreibung: electronic resource

Relation: https://doaj.org/toc/1551-0018

Zugangs-URL: https://doaj.org/article/6bbb842d0c624be59ef67ad20c5902f4