Non-Cartesian Self-Supervised Physics-Driven Deep Learning Reconstruction for Highly-Accelerated Multi-Echo Spiral fMRI [PDF]
Hongyi Gu +5 more
openalex +1 more source
Physical Intelligence in Small‐Scale Robots and Machines
“Physical intelligence” (PI) empowers biological organisms and artificial machines, especially at the small scales, to perceive, adapt, and even reshape their complex, dynamic, and unstructured operation environments. This review summarizes recent milestones and future directions of PI in small‐scale robots and machines.
Huyue Chen, Metin Sitti
wiley +1 more source
Self‐supervised multi‐view clustering in computer vision: A survey
In recent years, multi‐view clustering (MVC) has had significant implications in the fields of cross‐modal representation learning and data‐driven decision‐making.
Jiatai Wang +5 more
doaj +1 more source
Seismic Blind Deconvolution Based on Self-Supervised Machine Learning
Seismic deconvolution is a useful tool in seismic data processing. Classical non-machine learning deconvolution methods usually apply quite a few constraints to both wavelet inversion and reflectivity inversion.
Xia Yin +3 more
doaj +1 more source
Magnetic Domain Texture in Fe3O4 Thin Films on SiO2 Nanospheres
Fe3O4 thin films grown on ordered SiO2 nanospheres form curved nanocaps with 3D geometry, inducing magnetic domain texture. X‐ray spectromicroscopy (XMCD‐PEEM), cross‐sectional electron microscopy (STEM), and polarized grazing‐incidence small‐angle neutron scattering (GISANS) reveal how topography modulates magnetization.
Mai Hussein Hamed +14 more
wiley +1 more source
BYOL-S: Learning Self-supervised Speech Representations by Bootstrapping [PDF]
Gasser Elbanna +5 more
openalex +1 more source
Adversarial Self-Supervised Contrastive Learning
Existing adversarial learning approaches mostly use class labels to generate adversarial samples that lead to incorrect predictions, which are then used to augment the training of the model for improved robustness. While some recent works propose semi-supervised adversarial learning methods that utilize unlabeled data, they still require class labels ...
Kim, Minseon +2 more
openaire +2 more sources
A Wearable, Dual Closed‐loop Insulin Delivery System for Precision Diabetes Management
A wearable dual closed‐loop drug delivery system for diabetes management is presented, featuring an adaptive, on‐body trained edge AI algorithm for precise insulin delivery, which enhances therapeutic accuracy from 92.10% to 98.82%. Abstract Effective blood glucose management is an increasing demand worldwide.
Xuecheng He +12 more
wiley +1 more source
Self-supervised Learning for Single View Depth and Surface Normal Estimation [PDF]
Huangying Zhan +3 more
openalex +1 more source
Towards Learning Discrete Representations via Self-Supervision for Wearables-Based Human Activity Recognition [PDF]
Harish Haresamudram +2 more
openalex +1 more source

