Results 141 to 150 of about 31,077 (302)

Structured Pruning Learns Compact and Accurate Models

, 2022
The growing size of neural language models has led to increased attention in model compression. The two predominant approaches are pruning, which gradually removes weights from a pre-trained model, and distillation, which trains a smaller compact model ...
Zhong, Zexuan, Xia, Mengzhou, Chen, Danqi   +2 more
core  

Interface‐Engineered Li–Ni–O/Ni Heterostructures With Tunable Vacancies on Biochar Enable Biomass‐to‐Hydrogen Conversion

Carbon Energy, EarlyView.
This study shows that lithium induces Li–Ni–O/Ni heterostructures on biochar, creating oxygen vacancies and optimal acid–base environments that enhance steam activation and suppress carbon deposition. These interfacial effects and synergistic modulation lead to high hydrogen yields at moderate temperatures, providing a scalable catalyst design strategy
Jiali Guo, Jit Loong Cyrus Foo, Guicai Liu, Liya Ge, Andrei Veksha, Wei Ping Chan, Dean Song, Grzegorz Lisak   +7 more
wiley   +1 more source

Improvement of the Decision Support System Model for College Students' Entrepreneurship Path Planning Based on Tree Pruning Algorithm

Engineering Reports
Aiming at the problems of high decision‐making difficulty and insufficient adaptability of traditional algorithms in college students' entrepreneurship path planning, this paper proposes a decision support system model for college students ...
Lu Gao, Renjia Zhang
doaj   +1 more source

Inter‐Verified Binary Classification Models: Unveiling Key Polymer Features Against Staphylococcus aureus

Carbon Innovation, EarlyView.
ABSTRACT Antimicrobial peptides (AMPs) are promising alternatives to antibiotics but require precise and costly sequence control for efficacy. Synthetic polymers composed of acrylate and acrylamide offer a simpler and more time‐efficient antibacterial strategy compared to AMPs.
Jooyoung Roh, Cyrille Boyer, Priyank Kumar   +2 more
wiley   +1 more source

Safety soft sensor development for pilot‐scale ilmenite electric arc furnace using long short‐term memory‐based architecture

The Canadian Journal of Chemical Engineering, EarlyView.
Abstract Ilmenite electric arc furnaces (EAFs) are used for smelting titanium‐iron oxide ore at high temperatures generated by electrical arcs to produce titanium slag and pig iron. As these units are pushed to their limits, ensuring safe and reliable operation becomes challenging.
Antony Gareau‐Lajoie, Djogap F. Chrysler J., Daniel Rodrigues, Marie‐Ève Gosselin, Moncef Chioua   +4 more
wiley   +1 more source

Harnessing machine learning and optimization for informed chemical engineering decisions: A styrene reactor analysis

The Canadian Journal of Chemical Engineering, EarlyView.
This study shows that integrating multiple machine learning models with optimization and decision‐making improves chemical process design, and that a consensus‐based strategy across models provides more robust and reliable operating recommendations than any single model, especially under limited or noisy data conditions.
Farough Agin, Clémence Fauteux‐Lefebvre, Jules Thibault   +2 more
wiley   +1 more source

Pre‐industrial land‐use limits contemporary shrub encroachment in the French Alps

Ecography, EarlyView.
Shrub encroachment has become a global phenomenon in recent decades. While global warming in the Arctic is often cited as the primary cause, human‐managed mountain regions have experienced intense historical land‐use that may also play a considerable role.
Baptiste Nicoud, Florian Carrière, Arthur Bayle, Christian Devlieger, Philippe Choler, Christophe Corona, Sandra Lavorel   +6 more
wiley   +1 more source

Decision Tree Pruning with Privacy-Preserving Strategies

Electronics
Machine learning techniques, particularly decision trees, have been extensively utilized in Network-based Intrusion Detection Systems (NIDSs) due to their transparent, rule-based structures that enable straightforward interpretation. However, this transparency presents privacy risks, as decision trees may inadvertently expose sensitive information such
Yee Jian Chew, Shih Yin Ooi, Ying Han Pang, Zheng You Lim   +3 more
openaire   +1 more source

Climatic drivers prevail in montane and lowland Odonata latitudinal diversity gradients, but human modification erodes lowland patterns

Ecography, EarlyView.
Latitudinal diversity gradients (LDGs) arise from the interplay of historical, ecological, and evolutionary processes, yet these drivers may differ across landforms. Mountains, with steep elevational and climatic gradients, often sustain distinct diversity dynamics compared with adjacent lowlands, where vertical climatic gradients are weak and human ...
Zhenyuan Liu, Bo‐Ping Han, Haomiao Zhang, Ramiro Martín‐Devasa, Jens‐Christian Svenning, Janne Soininen   +5 more
wiley   +1 more source

Macroecological relationships of ant diversity with increasing aridity in Australian tropical savannas: contrasting responses of epigaeic and hypogaeic assemblages

Ecography, EarlyView.
Studies using climatic gradients play a key role in our understanding of the importance of rainfall and temperature as factors regulating species diversity and distribution, and thus of likely responses to climate change. However, such studies currently consider above‐ground species only, ignoring the diverse hypogaeic (subterranean) invertebrate fauna.
François Brassard, Giovanny Destour, Tristan Klaftenberger, Alan N. Andersen   +3 more
wiley   +1 more source