Results 91 to 100 of about 6,929,542 (320)

Optimization of Annealed Importance Sampling Hyperparameters

, 2023
AbstractAnnealed Importance Sampling (AIS) is a popular algorithm used to estimates the intractable marginal likelihood of deep generative models. Although AIS is guaranteed to provide unbiased estimate for any set of hyperparameters, the common implementations rely on simple heuristics such as the geometric average bridging distributions between ...
Shirin Goshtasbpour, Fernando Pérez-Cruz   +1 more
openaire   +4 more sources

Symbolic Regression and Multi‐Objective Optimization of the Flory–Huggins Interaction Parameter for Hydrogels

Advanced Engineering Materials, EarlyView.
We develop a data‐driven method to derive the mathematical expressions of the Flory–Huggins interaction parameter χ for the swelling behavior of temperature–responsive hydrogels. Starting from initial assumptions of χ, our workflow combines Bayesian optimization, Flory–Rehner theory, and symbolic regression to generate candidate χ expressions.
Yawen Wang, Adrian Ehrenhofer, Thomas Wallmersperger   +2 more
wiley   +1 more source

Generalized Population-Based Training for Hyperparameter Optimization in Reinforcement Learning [PDF]

IEEE Transactions on Emerging Topics in Computational Intelligence
Hyperparameter optimization plays a key role in the machine learning domain. Its significance is especially pronounced in reinforcement learning (RL), where agents continuously interact with and adapt to their environments, requiring dynamic adjustments ...
Hui Bai, Ran Cheng
semanticscholar   +1 more source

Neighbor Regularized Bayesian Optimization for Hyperparameter Optimization

CoRR, 2022
Bayesian Optimization (BO) is a common solution to search optimal hyperparameters based on sample observations of a machine learning model. Existing BO algorithms could converge slowly even collapse when the potential observation noise misdirects the optimization.
Lei Cui, Yangguang Li 0001, Xin Lu, Dong An 0002, Fenggang Liu   +4 more
openaire   +4 more sources

Machine Learning‐Supported Analysis for Predicting and Visualizing Nonlinear Relationships Between Material Properties in Electroplated Chromium Layers

Advanced Engineering Materials, EarlyView.
This study applies machine learning regression to predict chromium layer thickness in decorative trivalent chromium electroplating, using 441 experiments from laboratory‐scale (1L) and pilot‐scale (14L) setups. Tree‐based models, particularly CatBoost, outperformed linear regression by capturing nonlinear parameter interactions (R2$R^2$ up to 0.77 ...
Christoph Baumer, Christopher Mai, Luca Eisentraut, Andreas Bund, Ricardo Buettner   +4 more
wiley   +1 more source

BAYESIAN OPTIMIZATION FOR TUNING HYPERPARAMETRS OF MACHINE LEARNING MODELS: A PERFORMANCE ANALYSIS IN XGBOOST

Компютерні системи та інформаційні технології
The performance of machine learning models depends on the selection and tuning of hyperparameters. As a widely used gradient boosting method, XGBoost relies on optimal hyperparameter configurations to balance model complexity, prevent overfitting, and ...
Микола ЗЛОБІН, Володимир БАЗИЛЕВИЧ   +1 more
doaj   +1 more source

Hyperparameter Optimization Across Problem Tasks [PDF]

, 2019
Hyperparameter Optimization is a task that is generally hard to accomplish as the correct setting of hyperparameters cannot be learned from the data directly.
Schmidt-Thieme, Lars, Schilling, Nicolas, Windler, Torben   +2 more
core   +1 more source

A Practical Noise2Noise Denoising Pipeline for High‐Throughput Raman Spectroscopy

Advanced Engineering Materials, EarlyView.
A lightweight and reproducible denoising pipeline for high‐throughput Raman spectroscopy is introduced, based on a 1D convolutional autoencoder trained with a Noise2Noise strategy. Using only repeated short‐exposure acquisitions, the method suppresses stochastic noise without reference spectra, enabling reliable spectral reconstruction while preserving
David Martin‐Calle, Cesar Alvarez‐Llamas, Vincent Motto‐Ros, Christophe Dujardin, Jérémie Margueritat, David Rodney   +5 more
wiley   +1 more source

New AI‐Assisted Approach for Expanding the Solution Space: Application to Lattice Structure Design

Advanced Engineering Materials, EarlyView.
This work introduces an innovative framework for designing structured materials by ex panding the design space through reparameterization of qualitative variables into continuous structural descriptors. Combined with machine‐learning‐based prediction and multi‐objective optimization, the approach enables the discovery of novel lattice architectures ...
G. H. Gahimbare, Y. Beideler, D. Chune, R. De Guio, T. Roland, H. Chibane   +5 more
wiley   +1 more source

Optimizing Hyperparameters with Conformal Quantile Regression

CoRR, 2023
Many state-of-the-art hyperparameter optimization (HPO) algorithms rely on model-based optimizers that learn surrogate models of the target function to guide the search. Gaussian processes are the de facto surrogate model due to their ability to capture uncertainty but they make strong assumptions about the observation noise, which might not be ...
David Salinas, Jacek Golebiowski, Aaron Klein, Matthias W. Seeger, Cédric Archambeau   +4 more
openaire   +4 more sources