Results 101 to 110 of about 6,466,948 (249)
Probabilistic Safety for Bayesian Neural Networks
We study probabilistic safety for BayesianNeural Networks (BNNs) under adversarial in-put perturbations. Given a compact set of input points,T⊆Rm, we study the probability w.r.t. the BNN posterior that all the pointsinTare mapped to the same region S in theoutput space.
Matthew Wicker +3 more
openaire +5 more sources
On the shape of posterior densities and credible sets in instrumental variable regression models with reduced rank: an application of flexible sampling methods using neural networks [PDF]
Likelihoods and posteriors of instrumental variable regression models with strongendogeneity and/or weak instruments may exhibit rather non-elliptical contours inthe parameter space.
Hoogerheide, L.F. +2 more
core
A sea turtle‐inspired amphibious robot uses simulation‐guided optimization to adapt its flipper coordination to different environments. Sensitivity analysis and multiobjective Bayesian optimization identify terrain‐dependent controllers that are validated on hardware.
Nnamdi C. Chikere, Yasemin Ozkan‐Aydin
wiley +1 more source
Linear models, smooth transition autoregressions and neural networks for forecasting macroeconomic time series: A reexamination [PDF]
In this paper we examine the forecast accuracy of linear autoregressive, smooth transition autoregressive (STAR), and neural network (NN) time series models for 47 monthly macroeconomic variables of the G7 economies.
Timo Teräsvirta +2 more
core
Output Partitioning of Neural Networks [PDF]
Many constructive learning algorithms have been proposed to find an appropriate network structure for a classification problem automatically. Constructive learning algorithms have drawbacks especially when used for complex tasks and modular approaches ...
Yinan, Q. +11 more
core +1 more source
Robotics in Ionizing Radiation Environments: Vulnerabilities, Hardening, and Future Directions
Radiation exposure causes cumulative degradation in robotic sensors, actuators, and electronic logic, often leading to undetected failures. This study maps component vulnerabilities, evaluates hardening and shielding strategies, and identifies testing gaps that must be addressed to enable next‐generation radiation‐tolerant robotic systems.
Francis A. K. Djirackor +4 more
wiley +1 more source
Adaptive Motion Scaling With Phase‐and‐Gaze Awareness for Laparoscopic Robotic Teleoperation
The proposed adaptive motion scaling framework for laparoscopic robotic teleoperation integrates real‐time phase recognition, robot kinematics, and eye‐tracking. Bayesian optimization enables personalized tuning. The inferred phase probabilities and dominance weights suppress scaling fluctuations during fine manipulation while preserving responsiveness
Jingsong Liu +5 more
wiley +1 more source
Machine learning has evolved into a potent tool for analysing patterns and making predictions from complex data. In this machine learning era, we employed neural network techniques to estimate the parameters of statistical distributions.
P. T. Amrutha, C. S. Rajitha
doaj +1 more source
Physics‐Grounded Rapid Grasp Pose Synthesis for Novel Objects
Grasping models trained on large datasets often fail on unfamiliar objects. A fast pipeline reconstructs each new object with NeRF, samples force‐closure grasps six times faster than antipodal sampling, and removes unstable grasps through GPU‐accelerated physics simulation.
Xiao Hu, Yang Ye
wiley +1 more source
Accurate inverse solution of process parameters by surface roughness is crucial for precision gear grinding processes. When inversely solving process parameters, model parameters are typically obtained by fitting experimental data.
Qi Zhang +5 more
doaj +1 more source

