A Generative Neuro‐Symbolic AI for Protein Sequence Design
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
An antiferroelectric tunnel junction serves as a reservoir computing node, where field‐induced phase transitions and spontaneous ZrO2 relaxation deliver nonlinear fading memory. An In‐Ga‐Zn oxide interlayer enlarges the memory margin and multibit state richness.
Taegyu Kwon +13 more
wiley +1 more source
Probing Electrocatalyst Design for Product Selectivity in CO2 Reduction
Selective CO2 electroreduction is controlled by catalyst structure, dynamic surface reconstruction, adsorbate interactions, and reaction microenvironment. This review summarizes Cu‐ and non‐Cu‐based catalysts, single‐atom and molecular systems, degradation pathways, and operando/computational strategies for steering CO2RR toward hydrocarbons ...
Suvodeep Sen +5 more
wiley +1 more source
BSc-DARTS: Neural Architecture Search for Automatic Diagnosis of Bone Metastases in Bone Scintigrams [PDF]
Cao Y +7 more
europepmc +1 more source
Review of Gastrointestinal Capsule Robots: Materials, Actuation, Diagnostics, and Therapeutics
This review examines gastrointestinal capsule robots through a systems engineering lens, covering shell materials, actuation and localization strategies, diagnostic modalities, therapeutic interventions, and commercialization status. By mapping the complete technological chain from fabrication to clinical deployment, this work identifies critical gaps ...
Zengsheng Li +8 more
wiley +1 more source
COOT-CNN: joint architecture and training-strategy optimization for binary colorectal histology patch classification. [PDF]
Raza A, Ahmed AET, Kentour M, Hanif F.
europepmc +1 more source
Physics‐Aware Machine‐Learning‐Driven Inverse Design of Broadband Ultra‐Open Acoustic Metamaterials
A physics‐aware machine‐learning framework enables inverse design of ultra‐open acoustic silencers by decoupling spectral and radial design spaces. The approach rapidly identifies broadband, compact, and highly ventilated architectures, while revealing hidden linear design rules that link geometry, impedance matching, and acoustic performance.
Zhiwei Yang +5 more
wiley +1 more source
Research on deep learning architecture optimization method for intelligent scheduling of structural space. [PDF]
Ying W, Hui L.
europepmc +1 more source
Optimizing deep learning models for on-orbit deployment through neural architecture search. [PDF]
Del Prete R +6 more
europepmc +1 more source
Artificial Intelligence‐Guided Phenotypic Drug Repurposing Against Streptococcus pneumoniae
Integrated artificial intelligence (AI) ensembles identify highly potent repurposed drugs against Streptococcus pneumoniae. Prospective virtual screening of 6747 drugs uncovers nine active candidates, including thiostrepton. This drug exhibits low picomolar potency and bypasses established resistance in multidrug‐resistant strains, demonstrating an ...
Joshua S. Fitch +6 more
wiley +1 more source

