Adopting physics-based models, feature importance studies, or techniques like SHAP in conjunction with prediction algorithms are some solutions to this issue. Once more, these are some of the topics that require further investigation. Although many complicated engineering problems have answers that are ...
Evolutionary Algorithms (EAs), including Evolutionary Strategies (ES) and Genetic Algorithms (GAs), have been widely accepted as competitive alternatives t
values of the coefficients) and L2 (the sum of the squared values of the coefficients) penalties, elastic net addresses the limitations of alternative linear models such as LASSO regression (not capable of handling multi-collinearity) and Ridge Regression (may not produce sparse-enough solutions)70...
Recent research uses deep learning to propose solutions for these and related issues. However, deep learning faces problems like overfitting that may undermine the effectiveness of its applications in solving different network problems. This paper considers the overfitting problem of convolutional neural ...
2.1. Proposed methods for combining first principles and machine learning 2.1.1. Introduction to the proposed methods Up to now, in machine learning Virtual Flow Metering solutions reported in the literature, the measurements of pressures and temperatures available in the system have been used directly...
Next, a machine learning technique was applied to establish a model between the input data and determining parameters of a decline curve analysis model by fitting the generated cumulative production rate. Overall coefficients of determination (R2) of the three Arps decline curve factors were 0.966, ...
As a branch of RL, the deep Q-learning network (DQN) combines the perception ability of deep learning and its own decision-making ability (Mnih et al., 2013) and can supply new solutions to cognitive decision-making problems in complex states. As a new method in computer science, RL has...
Machine-learning control is of particular relevance to our work, where a machine-learning algorithm is applied to control a complex system and generate an effective control law that maps the desired system output to the input. More specifically, for complex control problems where an accurate model...
{M}\), see Table2.1, but this is not the canonical link of the selected models. In the gamma GLM this leads to a convex minimization problem, but in Tweedie’s GLM withp = 2.5 and in the inverse Gaussian GLM we have non-convex minimization problems, see Example5.6. Therefore, ...
There is therefore no backwards feedback for solving inverse problems such as geometry optimisation. This shortcoming results in manual trial-and-error implementations in industrial settings, which could miss potential optima, particularly for complicated geometries. Therefore, there is a need for a ...