Results 101 to 110 of about 28,332 (249)

ORTHOGONAL PERMUTATION SAMPLING FOR SHAPLEY VALUES: UNBIASED STRATIFIED ESTIMATORS WITH VARIANCE GUARANTEES

Journal of Intelligent Systems and Applied Data Science
Shapley values for feature attribution suffer from high variance requiring thousands of model evaluations. We introduce Orthogonal Permutation Sampling (OPS), achieving provable variance reduction through: (i) exact position stratification, (ii ...
YASH VARSHNEY, RANAV TYAGI, ANURAG SINHA
doaj   +1 more source

An Autonomous Large Language Model‐Agent Framework for Transparent and Local Time Series Forecasting

Advanced Intelligent Discovery, EarlyView.
Architecture of the proposed large language model (LLM)‐based agent framework for autonomous time series forecasting in thermal power generation systems. The framework operates through a vertical pipeline initiated by natural language queries from users, which are processed by the LLM Agent Core powered by Llama.cpp and a ReAct loop with persistent ...
William Gouvêa Buratto, Gabriel Villarrubia Gonzalez, Ademir Nied, Laio Oriel Seman, Anne Carolina Rodrigues Klaar, Stefano Frizzo Stefenon   +5 more
wiley   +1 more source

Conditional Feature Selection: Evaluating Model Averaging When Selecting Features with Shapley Values

Geomatics
In the field of geomatics, artificial intelligence (AI) and especially machine learning (ML) are rapidly transforming the field of geomatics with respect to collecting, managing, and analyzing spatial data.
Florian Huber, Volker Steinhage
doaj   +1 more source

AI‐Driven Cancer Multi‐Omics: A Review From the Data Pipeline Perspective

Advanced Intelligent Discovery, EarlyView.
The exponential growth of cancer multi‐omics data brings opportunities and challenges for precision oncology. This review systematically examines AI's role in addressing these challenges, covering generative models, integration architectures, Explainable AI for clinical trust, clinical applications, and key directions for clinical translation.
Shilong Liu, Shunxiang Li, Kun Qian
wiley   +1 more source

AS‐pHopt: An Optimal pH Prediction Model Enhanced by Active Site of Enzymes

Advanced Intelligent Discovery, EarlyView.
To address the low accuracy of enzyme optimal pH (pHopt) prediction, this study develops active site‐based pHopt (AS‐pHopt), a prediction model enhanced by active site information and pseudo‐label prediction. Integrating key structural and physicochemical features affecting enzyme pHopt, AS‐pHopt uses Evolutionary Scale Modeling (ESM)‐2 with active ...
Wenxiang Song, Xinmin Li, Hongbo Yu, Le Xiong, Guixia Liu, Weihua Li, Yun Tang   +6 more
wiley   +1 more source

Calculation of exact Shapley values for explaining support vector machine models using the radial basis function kernel. [PDF]

Sci Rep, 2023
Mastropietro A, Feldmann C, Bajorath J.
europepmc   +1 more source

Deep Learning–Based Extraction of Promising Material Groups and Common Features from High‐Dimensional Data: A Case of Optical Spectra of Inorganic Crystals

Advanced Intelligent Discovery, EarlyView.
We report a novel interpretation method for deep learning models based on feature extraction and clustering. Applying this method to an atomistic line graph neural network (ALIGNN) model trained on optical absorption spectra of 2,681 inorganic compounds obtained from first‐principles calculations, we successfully identify key factors underlying ...
Akira Takahashi, Yu Kumagai, Arata Takamatsu, Fumiyasu Oba   +3 more
wiley   +1 more source

Calculation of exact Shapley values for support vector machines with Tanimoto kernel enables model interpretation. [PDF]

iScience, 2022
Feldmann C, Bajorath J.
europepmc   +1 more source

GraphX-Net: A Graph Neural Network-Based Shapley Values for Predicting Breast Cancer Occurrence

IEEE Access
Breast cancer is a major health problem worldwide, and an accurate prediction of its recurrence is crucial to early detection of recurrence and personalized treatment.
Abdullah Basaad, Shadi Basurra, Edlira Vakaj, Mohammed Aleskandarany, Mohammed M. Abdelsamea   +4 more
doaj   +1 more source

Interpretable Machine Learning for Bandgap Prediction and Descriptor‐Guided Design Rules of Phosphates

Advanced Intelligent Discovery, EarlyView.
An explainable CatBoost model was trained to predict the bandgaps of 474 phosphate crystals based on composition and density descriptors. SHAP analysis identified two key variables—d‐electron‐count dispersion and atomic‐density dispersion—as the primary drivers of the model's predictions.
Wenhu Wang, Abudukadi Tudi, Ran An, Zhihua Yang   +3 more
wiley   +1 more source