Advancements in Optimization and Machine Learning for Complex Systems

The field of optimization and machine learning is witnessing significant advancements, particularly in the development of novel algorithms and techniques for complex systems. Recent studies have focused on improving the efficiency and accuracy of optimization methods, such as genetic algorithms and evolutionary metaheuristics, for applications in disease prediction, power electronics, and numerical integration. Additionally, the integration of machine learning techniques, such as neural networks and surrogate models, has shown promising results in accelerating simulation and optimization processes. Notably, the use of physics-informed neural networks and probabilistic learning-based stochastic surrogate models has demonstrated potential in simulating complex systems and optimizing power electronic converters. Furthermore, the development of optimized gate merging methods and dynamic heuristic approaches has mitigated issues in digital circuits and improved computational efficiency. Overall, these advancements are paving the way for more efficient and accurate solutions to complex problems in various fields.

Noteworthy papers include:

  • A study on a multiprocessing interface genetic algorithm for optimizing a multilayer perceptron for disease prediction, which achieved high accuracy scores of up to 100% on certain datasets.
  • A paper on a self-adaptive IMEX time integration scheme, which demonstrated unconditional stability and high-fidelity accuracy in numerical experiments.
  • A work on a physics-informed bidirectional long-short term memory neural network model for simulating a closed-loop dc-dc converter, which outperformed other methods in terms of accuracy and consistency.

Sources

Development of a Multiprocessing Interface Genetic Algorithm for Optimising a Multilayer Perceptron for Disease Prediction

A Study of Hybrid and Evolutionary Metaheuristics for Single Hidden Layer Feedforward Neural Network Architecture

IMEX-RB: a self-adaptive IMEX time integration scheme exploiting the RB method

Simulation of a closed-loop dc-dc converter using a physics-informed neural network-based model

ConStellaration: A dataset of QI-like stellarator plasma boundaries and optimization benchmarks

Training Flexible Models of Genetic Variant Effects from Functional Annotations using Accelerated Linear Algebra

Surrogate-Assisted Evolution for Efficient Multi-branch Connection Design in Deep Neural Networks

Optimal Parameter Design for Power Electronic Converters Using a Probabilistic Learning-Based Stochastic Surrogate Model

Optimising 4th-Order Runge-Kutta Methods: A Dynamic Heuristic Approach for Efficiency and Low Storage

OptGM: An Optimized Gate Merging Method to Mitigate NBTI in Digital Circuits

Built with on top of