Results 81 to 90 of about 14,357 (268)
Advanced Science, EarlyView.SPADE integrates spatial transcriptomics with single‐cell RNA sequencing by using cell–cell communications (CCC) as a guide for spatial mapping. It improves cell‐type localization, enhances sparse gene‐expression signals, and reveals CCC programs at single‐spot resolution.
Xinyi Li, Ning Zhang, Zijie Jin
wiley +1 more source
CauFinder: Steering Cell‐State and Phenotype Transitions by Causal Disentanglement Learning
Advanced Science, EarlyView.CauFinder combines causal disentanglement modeling and network control to prioritize causal drivers of cell‐state transitions from observational transcriptomic data. The framework separates transition‐relevant signals from spurious associations, nominates intervention targets across biological and disease contexts, and identifies DAAM1 as an actionable
Chengming Zhang +11 more
wiley +1 more source
CoRR, 2018 In this paper, we describe the "implicit autoencoder" (IAE), a generative autoencoder in which both the generative path and the recognition path are parametrized by implicit distributions. We use two generative adversarial networks to define the reconstruction and the regularization cost functions of the implicit autoencoder, and derive the learning ...
openaire +2 more sources
Advanced Science, EarlyView.Decoupling biological signals from unwanted variation in multi‑condition single‑cell RNA sequencing data remains challenging. CAPER disentangles condition‑associated biological effects from sample heterogeneity through matrix factorization, producing interpretable latent factors and a batch‑corrected expression matrix.
Ye Li +6 more
wiley +1 more source
A Generative Neuro‐Symbolic AI for Protein Sequence Design
Advanced Science, EarlyView.We introduce EffieDes, a neuro‐symbolic framework coupling deep learning‐based fitness landscape parameterization with exact automated reasoning. Unlike greedy sampling, EffieDes identifies sequences that globally optimize fitness while satisfying intricate design constraints.
Marianne Defresne +12 more
wiley +1 more source
StackingNet: Collective Inference Across Independent AI Foundation Models
Advanced Science, EarlyView.ABSTRACT Artificial intelligence (AI) built on large foundation models has transformed language understanding, computer vision, and reasoning, yet these systems remain isolated and cannot readily share their capabilities. Coordinating the complementary strengths of independently developed, black‐box foundation models is essential for trustworthy ...
Siyang Li +4 more
wiley +1 more source
Composite Denoising Autoencoders [PDF]
, 2016 In representation learning, it is often desirable to learn features at different levels of scale. For example, in image data, some edges will span only a few pixels, whereas others will span a large portion of the image. We introduce an unsupervised representation learning method called a composite denoising autoencoder CDA to address this.
Geras, Krzysztof, Sutton, Charles
openaire +1 more source
Advanced Science, EarlyView.A soft hybrid multi‐wavelength PPG wearable acquires neonatal signals. Synchronized PPG and invasive ABP data are segmented into fixed windows. A 1D‐EfficientNet model predicts segment‐level SBP and DBP. Model performance is examined with retrospective subgroup analysis across acquisition conditions.
Wenqi Shi +12 more
wiley +1 more source
Advanced Science, EarlyView.A latent diffusion‐based framework is proposed for designing functionally graded metamaterials with perfect connectivity. By integrating vector‐quantized latent representations with mechanistic guidance, the framework enables accurate inverse design toward target elastic properties.
Jongbin Yu, Dosung Lee, Namjung Kim
wiley +1 more source