Results 111 to 120 of about 60,311 (308)
Advanced Science, EarlyView.Non‐canonical amino acids (ncAAs) enhance peptide therapeutics but remain difficult to model computationally. SinCAA, a similarity‐enhanced pretraining framework, jointly optimizes contrastive learning guided by a novel conformational similarity metric with masked node reconstruction, capturing both functional relationships and chemical identity of ...
Chencheng Xu +8 more
wiley +1 more source
Future InternetThe rapid expansion of network environments has introduced significant cybersecurity challenges, particularly in handling high-dimensional traffic and detecting sophisticated threats.
Anubhav Kumar +5 more
doaj +1 more source
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 +5 more
wiley +1 more source
Pretraining procedure for autoencoder.
, 2016 The weights of a 5-layer autoencoder (right) are copied from two 3-layer autoencoders (left and middle).
Takamichi Nakamoto (3165207) +1 more
core +1 more source
Uncertainty Quantification of Autoencoder-based Koopman Operator [PDF]
, 2023 This paper proposes a method for uncertainty quantification of an autoencoder-based Koopman operator. The main challenge of using the Koopman operator is to design the basis functions for lifting the state.
Quan, Ying Shuai +2 more
core
Supervised temporal autoencoder for stock return time-series forecasting
, 2021 Financial markets are noisy learning environments. We propose an approach that regularizes the Temporal Convolutional Network using a supervised autoencoder, which we term the Supervised Temporal Autoencoder (STAE).
Wong, SYK +7 more
core +1 more source
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 +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 +15 more
wiley +1 more source
Image compression module based neural network autoencoders
, 2023 A new approach is proposed to data compression in the form of a neural network module based on the structure of autoencoders, which has the most optimal learning time, compression level and obtains sufficiently clear image reconstruction.
Lesyk, V.O., Doroshenko, А.Yu.
core
qcGEM: A Graph‐Based Molecular Representation With Quantum Chemistry Awareness
Advanced Science, EarlyView.In this work, we introduce qcGEM, a quantum‐chemistry‐aware graph‐based embedding of molecules that incorporates physical priors into molecular representation learning. Through a specially designed model architecture to adequately process quantum chemistry information as well as three‐dimensional molecular geometry, qcGEM provides a compact, physics ...
Haoyu Wang +3 more
wiley +1 more source