Results 141 to 150 of about 4,772,239 (326)

Sparsity in Variational Autoencoders

CoRR, 2018
An Extended Abstract of this survey will be presented at the 1st International Conference on Advances in Signal Processing and Artificial Intelligence (ASPAI' 2019), 20-22 March 2019, Barcelona ...
openaire   +4 more sources

Predicting Single‐Cell Perturbation Responses Across Biological Contexts With a Deep Generative Model Integrating Optimal Transport

Advanced Science, EarlyView.
Single‐cell perturbation responses are predicted across held‐out biological contexts using scPILOT, a query‐conditioned two‐stage latent response‐transfer framework. A shared latent representation supports cell‐level response estimation by latent optimal transport, followed by Leiden‐localized query‐specific transfer and adaptive weighting.
Jialiang Wang, Ziqi Liu, Zhengqian Zhang, Yikun Cao, Junjun Ren, Peng Cheng, Jingjing Tian, Lingyun Xie, Xin Lu, Zhanwei Du, Yongzhuang Liu   +10 more
wiley   +1 more source

Molecular Atlas of Key Food Odorants Reveals Mixture‐Level Organization and Enables Generative Aroma Design

Advanced Science, EarlyView.
KFO‐Atlas reveals how real‐world aromas are organized as structured mixtures rather than individual molecules. Building on these principles, KFO‐Gen, a generative AI framework, designs perceptually valid aroma formulations and reconstructs meat‐like aromas exclusively from plant‐derived odorants, providing a foundation for mixture‐level studies and AI ...
Jingzhi Zhang, Huadong Xing, Antonella Di Pizio, Qinfei Ke, Xingran Kou, Dachuan Zhang   +5 more
wiley   +1 more source

scGEN: Leveraging an Adaptive Weighting Strategy for Challenging Cell Clustering in Single‐Cell RNA‐seq

Advanced Science, EarlyView.
scGEN combines attribute–structure encoding, adaptive hard‐sample weighting, and regularized zero‐inflated negative binomial loss to resolve ambiguous single‐cell states. In the developing human pituitary, refined annotations reveal Pre.Gonado cells at 7 weeks and previously obscured developmental communication networks.
Fang Huang, Dayu Hu, Yaqi Dai, Aoyi Geng, Renxiang Guan, Hao Quan, Brett A. McGregor, Zhihan Wang, Nadeem Khan, Hongsheng Zhang, Jinghua Zhang, Mingyang Xia, Xiaoqiong Gu, Di Che, Junguk Hur, Xiaoyu Cui, Kai Guo   +16 more
wiley   +1 more source

A Facial Foundation Model for Clinical Biomarker Prediction and Real‐World Mobile Deployment

Advanced Science, EarlyView.
MedicalFaceFound is a facial foundation model developed through three‐stage progressive pre‐training, adapting representations from general vision to facial structure and clinical phenotypes. Using a single facial image, it estimates 62 biomarkers across multiple physiological systems and supports downstream risk prediction for cardiovascular ...
Tingfeng Xu, Huixuan Xu, Li Lin, Fei Wang, Yaodong Ding, Shuangqiao Liao, Xingyu Chen, Jichen Wen, Hongqiang Zhao, Jing Hao, Weixiang Zhao, Gong Zhang, Shiguang Shan, Yong Zeng, Hu Han, Minxian Wang   +15 more
wiley   +1 more source

A load forecasting method for IESs using TimeVAE and transfer learning

Zhejiang dianli
In this paper， an integrated learning approach based on time variational autoencoder （TimeVAE） and transfer learning （TL） is proposed to address the accuracy degradation in integrated energy systems （IESs） load forecasting caused by insufficient ...
CHEN Zhe, ZHOU Jinhui, JIN Donghui, CHEN Jiguang, MA Hengrui, ZHANG Jiaxin, ZHU Suxun   +6 more
doaj   +1 more source

Variational Methods for Nonsmooth Mechanics [PDF]

, 2003
In this thesis we investigate nonsmooth classical and continuum mechanics and its discretizations by means of variational numerical and geometric methods.
Fetecau, Razvan Constantin
core   +1 more source

Enhancing Specific Stiffness of Nano‐Architected Materials Through Generative AI and Active Learning

Advanced Science, EarlyView.
A structure‐centric generative framework searches voxelated design space directly, coupling a variational autoencoder with multi‐objective optimization and active learning to progressively expand the validated design domain. The discovered nano‐architectures, realized as pyrolytic carbon by two‐photon lithography, deliver 18–23% higher stiffness and 26–
Jinwook Yeo, Peter Serles, Donggeun Park, Julia R. Greer, Seunghwa Ryu   +4 more
wiley   +1 more source

Latent space improved masked reconstruction model for human skeleton-based action recognition

Frontiers in Neurorobotics
Human skeleton-based action recognition is an important task in the field of computer vision. In recent years, masked autoencoder (MAE) has been used in various fields due to its powerful self-supervised learning ability and has achieved good results in ...
Enqing Chen, Xueting Wang, Xin Guo, Ying Zhu, Dong Li   +4 more
doaj   +1 more source

Disentangling Disentanglement in Variational Autoencoders

, 2018
We develop a generalisation of disentanglement in VAEs---decomposition of the latent representation---characterising it as the fulfilment of two factors: a) the latent encodings of the data having an appropriate level of overlap, and b) the aggregate encoding of the data conforming to a desired structure, represented through the prior.
Mathieu, E, Rainforth, T, Siddharth, N, Teh, Y   +3 more
openaire   +4 more sources