Results 121 to 130 of about 1,262,485 (305)

Fine-tuning Transfer Learning based on DCGAN Integrated with Self-attention and Spectral Normalization for Bearing Fault Diagnosis

, 2023
Fine-tuning Transfer Learning based on DCGAN Integrated with Self-attention and Spectral Normalization for Bearing Fault ...
Samson Yu (13093116), Rui Yuan (615737), Yanan Wang (13100319), Yong Lv (5682287), Hongyu Zhong (2933082), Hieu Trinh (13073460)   +5 more
core   +1 more source

Influence of Si Content and Milling Duration on the Microstructure and Mechanical–Tribological Properties of AlCoCrFeNiSi High‐Entropy Alloys

Advanced Engineering Materials, EarlyView.
Si‐doped AlCoCrFeNi high‐entropy alloys are synthesized by mechanical alloying to reveal the effect of Si content and milling time on phase evolution, microstructural refinement, and tribological behavior. A transition from FCC to BCC structure, significant grain refinement, and enhanced hardness and wear resistance are achieved, with the 4 at% Si ...
Mustafa Okumuş, Mehmet Emin Demir, Sercan Özpınar   +2 more
wiley   +1 more source

Frozen Weights as Prior for Parameter-Efficient Fine-Tuning

IEEE Access
In the fields of natural language processing and computer vision, the emergence of large pre-trained models has led to the adoption of fine-tuning them for downstream tasks as an important paradigm. However, the full fine-tuning approach often comes with
Xiaolong Ma, Peishun Liu, Haojie Gao, Zikang Yan, Ningning Ma, Wenqiang Liu, Xuefang Wang, Ruichun Tang   +7 more
doaj   +1 more source

Tuning into the audible: an arts-based investigation into human connections with acoustic environments

, 2023
Tuning into the Audible is an arts-based research project probing human connections to the physicality and phenomena of sound-in-space. Sound shapes our spatial reality.
University of Lethbridge. Faculty of Fine Arts, Milinusic, Christina M.   +1 more
core  

Stretching the Printability Metric in Direct‐Ink Writing with Highly Extensible Yield‐Stress Fluids

Advanced Functional Materials, EarlyView.
This study introduces “drawability” as a new metric for assessing printability in direct‐ink writing, focusing on gap‐spanning performance and speed robustness. By designing yield‐stress fluids with high extensibility, we demonstrate that extensional strain‐to‐break significantly enhances printability.
Chaimongkol Saengow, Samya Sen, Joaquin Yus, Eliza E. Lovrich, Amanda G. Hoika, Kelly M. Chang, Arielle A. Pfeil, Nellie Haug, Amy J. Wagoner Johnson, Randy H. Ewoldt   +9 more
wiley   +1 more source

FedPFT: Federated Proxy Fine-Tuning of Foundation Models

Adapting Foundation Models (FMs) for down- stream tasks through Federated Learning (FL) emerges a promising strategy for protecting data privacy and valuable FMs.
Peng, Z, Wang, Z, Zhang, R, Wen, C, Chen, Y, Fan, X, Wang, C, Pan, S   +7 more
core   +1 more source

All‐in‐One Analog AI Hardware: On‐Chip Training and Inference with Conductive‐Metal‐Oxide/HfOx ReRAM Devices

Advanced Functional Materials, EarlyView.
An all‐in‐one analog AI accelerator is presented, enabling on‐chip training, weight retention, and long‐term inference acceleration. It leverages a BEOL‐integrated CMO/HfOx ReRAM array with low‐voltage operation (<1.5 V), multi‐bit capability over 32 states, low programming noise (10 nS), and near‐ideal weight transfer.
Donato Francesco Falcone, Victoria Clerico, Wooseok Choi, Tommaso Stecconi, Folkert Horst, Laura Bégon‐Lours, Matteo Galetta, Antonio La Porta, Nikhil Garg, Fabien Alibart, Bert Jan Offrein, Valeria Bragaglia   +11 more
wiley   +1 more source

Fine-Tuning and the Scope of Physical Laws

Zygon
Fine-tuning arguments claim that the precise parameter values and initial conditions required for complexity and life are extraordinarily improbable and demand explanation. This article examines whether such probability claims can be grounded in physics.
Chris Smeenk
doaj   +2 more sources

Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples

With the evolution of self-supervised learning, the pre-training paradigm has emerged as a predominant solution within the deep learning landscape.
Wan, W, Yao, D, Li, M, Jin, H, Liu, W, Zhang, LY, Zhou, Z, Xue, L, Zhang, Y, Hu, S   +9 more
core   +1 more source

Laser‐Induced Graphene from Waste Almond Shells

Advanced Functional Materials, EarlyView.
Almond shells, an abundant agricultural by‐product, are repurposed to create a fully bioderived almond shell/chitosan composite (ASC) degradable in soil. ASC is converted into laser‐induced graphene (LIG) by laser scribing and proposed as a substrate for transient electronics.
Yulia Steksova, Anna Chiara Bressi, Marina Galliani, Attilio Marino, Gianni Ciofani, Eduardo Machado‐Charry, Hilda Gomez Bernal, Alessandra Francini, Luca Sebastiani, Francesco Greco   +9 more
wiley   +1 more source