Results 71 to 80 of about 22,594 (260)

Large Language Model‐Informed Dual‐Track AI Framework for the Synergistic Design of Crack‐Free and High‐Strength Superalloys

Advanced Science, EarlyView.
This paper illustrates a knowledge‐augmented dual‐track AI framework for advanced superalloy design. First, Large Language Models translate metallurgical heuristics into explicit rules to rapidly prune a vast compositional search space. Subsequently, LLM‐distilled priors safely guide a reinforcement learning agent during autonomous process optimization,
Jian Yao, Junkang Wu, Jie Su, Fuzhu Wang, Zi Wang, Li Wang, Liming Tan, Lan Huang, Feng Liu, Yong Liu   +9 more
wiley   +1 more source

CLAD: Criterion learner and attention distillation for automated CNN prunning

Journal of Automation and Intelligence
Filter pruning effectively compresses the neural network by reducing both its parameters and computational cost. Existing pruning methods typically rely on pre-designed pruning criteria to measure filter importance and remove those deemed unimportant ...
Zheng Li, Jiaxin Li, Shaojie Liu, Bo Zhao, Derong Liu   +4 more
doaj   +1 more source

Directional Pruning of Deep Neural Networks

CoRR, 2020
In the light of the fact that the stochastic gradient descent (SGD) often finds a flat minimum valley in the training loss, we propose a novel directional pruning method which searches for a sparse minimizer in or close to that flat region. The proposed pruning method does not require retraining or the expert knowledge on the sparsity level.
Shih-Kang Chao, Zhanyu Wang, Yue Xing 0002, Guang Cheng 0003   +3 more
openaire   +3 more sources

StackingNet: Collective Inference Across Independent AI Foundation Models

Advanced Science, EarlyView.
ABSTRACT Artificial intelligence (AI) built on large foundation models has transformed language understanding, computer vision, and reasoning, yet these systems remain isolated and cannot readily share their capabilities. Coordinating the complementary strengths of independently developed, black‐box foundation models is essential for trustworthy ...
Siyang Li, Chenhao Liu, Dongrui Wu, Zhigang Zeng, Lieyun Ding   +4 more
wiley   +1 more source

Efficient Pruning of Detection Transformer in Remote Sensing Using Ant Colony Evolutionary Pruning

Applied Sciences
This study mainly addresses the issues of an excessive model parameter count and computational complexity in Detection Transformer (DETR) for remote sensing object detection and similar neural networks.
Hailin Su, Haijiang Sun, Yongxian Zhao
doaj   +1 more source

A Panel of Circulating Exosomal sncRNAs Associated With Lung Cancer Risk up to 10 Years in Advance

Advanced Science, EarlyView.
Lung cancer is often diagnosed too late, and current screening overlooks many people at risk. In a long‐term study of smokers, a panel of small non‐coding RNAs carried in blood exosomes signals elevated lung cancer risk up to ten years before diagnosis, pointing toward a blood‐based tool for earlier risk detection.
Zhuokun Feng, Masaki Nasu, Ayman Abdul‐Ghani, Yu Chen, Shaoqiu Chen, Isam Mohd Ibrahim, Hanqiu Zhang, Lauren Higa, Zitong Gao, Hua Yang, Raphael Bueno, Loïc Le Marchand, Youping Deng   +12 more
wiley   +1 more source

CLRe: A Synergistic Dual‐Engine Framework for One‐Step Retrosynthesis Prediction

Advanced Science, EarlyView.
CLRe uses a contrastive difficulty score to order pretrained seq2seq fine‐tuning for retrosynthesis. Reaction embeddings define the ranking score, and a cumulative easy‐to‐hard schedule expands from the easiest subset to the full training set while earlier examples remain active.
Tianhao Su, Xitao Wang, Musen Li, Guanhua Qin, Shunbo Hu, Tong‐Yi Zhang   +5 more
wiley   +1 more source

Biochemically Constrained Multi‐Omics Integration Reveals Protein–Metabolite Dependencies Across Diseases

Advanced Science, EarlyView.
ProMetNet introduces a biologically constrained deep learning framework for proteo‐metabolomic integration by embedding Reactome‐derived pathway topology into neural networks. It captures non‐linear molecular dependencies and pathway‐level metabolic reorganization, enabling interpretable discrimination.
Minghui Zhao, Na Zhou, Ruotong Liu, Xiaofei Li, Jian Li, Fuzhong Xue, Qingzhen Hou   +6 more
wiley   +1 more source

Entropy-Guided Search Space Optimization for Efficient Neural Network Pruning

Algorithms
Neural network pruning is essential for deploying deep learning models on resource-constrained devices by reducing computational and memory demands.
Yicheng Qiu, Li Niu, Feng Sha, Zhaokun Cheng, Keiji Yanai   +4 more
doaj   +1 more source

Demystifying Neural Network Filter Pruning

CoRR, 2018
Based on filter magnitude ranking (e.g. L1 norm), conventional filter pruning methods for Convolutional Neural Networks (CNNs) have been proved with great effectiveness in computation load reduction. Although effective, these methods are rarely analyzed in a perspective of filter functionality.
Zhuwei Qin, Fuxun Yu, Chenchen Liu, Xiang Chen 0010   +3 more
openaire   +2 more sources