Weakly Supervised Semantic Segmentation (WSSS) is a challenging task in computer vision, as it relies on limited supervision to generate precise object localization maps, often using Class Activation Maps (CAMs). Traditional methods struggle with balancing localization accuracy and scalability due to their reliance on fixed network architectures and handcrafted strategies. Neural Architecture Search (NAS), despite its proven success in optimizing network designs across tasks, has not yet been explored in WSSS due to the need for efficient weight sharing. To address these limitations, we propose WEViT, a novel framework that integrates NAS with transformers to optimize network architectures and generate accurate and class-specific object localization maps for WSSS. Our approach leverages the weight entanglement strategy, enabling the supernet to train multiple subnets simultaneously while ensuring high-quality weight inheritance. This eliminates the need for retraining subnets from scratch, significantly reducing computational cost. The best-performing architecture, obtained through the evolutionary algorithm, is then utilized to extract attention weights from transformer heads. These weights are further refined using a Refinement Patch Affinity strategy, effectively removing background noise and enhancing focus on relevant classes in multi-class images. We also incorporate a regularization loss function during training to enhance the generation of class-discriminative localization maps, with experiments highlighting the critical role of transformer layer selection in this process. WEViT achieves state-of-the-art performance on PASCAL VOC 2012 and MS COCO, demonstrating the efficacy of applying NAS to WSSS for the first time and paving the way for scalable, efficient, and accurate segmentation solutions.

Enhancing chemical kinetic mechanism optimization using independent component analysis

Multi-threaded constraint programming-assisted space perception evolutionary algorithm for energy-efficient flexible job shop scheduling problem with sequence-dependent setup times

Most research on flexible job shop scheduling assumes constant processing speeds. However, in real production, machines need to operate at variable speeds to achieve energy-efficient scheduling, which requires balancing multiobjective between production efficiency and green development. Such tradeoffs thus trigger the phenomenon in which massive solutions converge to identical objective values (i.e., the multimodal property), which is often neglected in scheduling problems. To address the above challenges, this work introduces a knowledge-enhanced evolutionary multitasking memetic algorithm (KEMMA) to solve the multimodal multiobjective flexible job shop scheduling problem considering speed (MMFJSP-S). First, self-paced learning motivated us to construct a simple auxiliary task and employ an evolutionary multitasking (EMT) framework to tackle the complex MMFJSP-S. Moreover, a knowledge enhancement and explicit transfer strategy is designed to reduce the effects of negative transfer by reinforcing and sharing beneficial knowledge across tasks. Finally, a mapping transformation mechanism is proposed to handle the multimodal property of the MMFJSP-S in the decision space. By comparing with ten advanced algorithms, the experimental results verify the remarkable superiority of the proposed KEMMA in solving MMFJSP-S and reveal the significance of studying the multimodal property.

Optimization of a novel reverse vortex generator for thermal management in microwave electrothermal thrusters via evolutionary algorithms

A dual-population cooperative evolutionary algorithm based on contribution degree for large-scale many-objective optimization

Relation model-assisted multi-region evolutionary algorithm for expensive constrained optimization

A decision-support method for multi-manufacturer service network optimization: Leveraging shared service stations and a decomposition-based evolutionary algorithm

Optimizing lockdown schedules for emerging pandemics using multi-objective evolutionary algorithms

A two-stage robust rescheduling for flexible job shop based on release time prediction using a Q-learning-based hyper-heuristic evolutionary algorithm

An indicator-driven evolutionary algorithm for constrained multi-objective optimization with small feasible regions

Gate recurrent unit neural network inverse model prediction-based dynamic multi-objective evolutionary algorithm and application

A multi-objective robust optimization based on evolutionary algorithm for container routing problem under risks and uncertainties

IKUN: A mean-field game theoretic KD-tree density guided mechanism for evolutionary algorithms

Cooperative 4D trajectory planning for multi aerial vehicles combining receding horizon optimization and differential evolution algorithm in dynamic battlefield environment

Research on intelligent beam tuning for 2x3.0 MV tandem accelerator based on enhanced parameter adaptive differential evolution algorithm

Due to the increasing volume of freight traffic and the growing practice of structural monitoring for existing bridges, it has become crucial to exploit modal analysis to identify the magnitudes of axle loads and the frequency content of the structural response induced by vehicle crossings. In this paper, starting from the comparison between a simplified analytical model of the bridge and the relevant experimental responses, a multi-parameter identification method is proposed to identify the magnitude, the number, the axle distance, and the eccentricity of moving loads crossing the bridge. A bending-torsional beam model subjected to travelling loads characterized by non-uniform spacing and magnitudes, has been adopted to describe the bridge dynamics. The identification procedure is based on the Differential Evolution genetic algorithm. The proposed method is tested and validated using numerically simulated dynamic responses including the presence of noise. The main contribution of this work concerns the combined use of a beam model and the DE algorithm to identify heavy vehicle load distributions for skew road bridges. • A multi-parameter identification method is proposed to identify bridge moving loads. • The intensity, number, axle distance, and eccentricity of loads are identified. • The adopted 1-D beam model integrates bending and torsional responses. • Both load eccentricity and deck skewness can be considered. • Governing equations have been solved starting from a Faedo–Galerkin approach. • The differential evolution algorithm was adopted for the inverse problem. • An alternative convergence criterion (called “swarm criterion”) was proposed.

A dynamic neighborhood multi-objective differential evolution algorithm based on knee point preferences and its application in portfolio optimization

An ensemble of surrogate models helps improve the prediction quality and robustness of surrogate models, and in turn, the search performance of surrogate-assisted evolutionary algorithms (SAEAs). Although different degrees of smoothness of the approximated fitness landscapes need to be carefully designed for an effective ensemble, little attention has been paid to the explicit tuning of the degree of smoothness derived by surrogate models. This study proposes an adaptive ensemble SAEA, which automatically constructs plausible ensemble models by optimizing their parameter settings. Unlike existing adaptive/ensemble SAEAs, which consider prediction accuracy alone, the proposed algorithm optimizes the structure of radial basis function networks (RBFNs) by solving bi-objective minimization problems of approximation error and model complexity, resulting in robust ensemble models of accurate surrogate models with different degrees of smoothness of the approximated fitness landscapes. As a result, the over/under-fittings are reduced. Additionally, an infill criterion is designed so that surrogate models with different degrees of smoothness can contribute to the solution prescreening. The experimental results demonstrated the statistical superiority of our algorithm over state-of-the-art SAEAs on a single-objective benchmark and real-world problem sets under an expensive optimization scenario. The source code of the proposed algorithm is available at https://github.com/haranychan/EPOS.

Accurate prediction of thermal conductivity in (nano-PEG) composites is essential for accelerating thermal management material design. This study develops a hybrid Random Forest (RF) framework optimized using eight evolutionary algorithms, including (PSO), (GA), (WOA), (GWO), (CSA), (FPA), (FA), and (BA). A dataset of 229 experimental observations was used to model thermal conductivity as a function of temperature, PEG molecular weight, nanoparticle concentration, and nanoparticle form. Among evaluated models, the Bat Algorithm-optimized RF (RF-BA) achieved highest predictive efficiency with R2 = 0.995406, MSE = 0.000196, and AARE = 1.291%, while the PSO-optimized model (RF-PSO) demonstrated the fastest optimization runtime (96.9s) with competitive accuracy. Correlation and SHAP analyses revealed nanoparticle concentration as the dominant factor governing thermal conductivity (correlation coefficient = 0.75), followed by PEG molecular weight (0.56), temperature (0.33), and nanoparticle form (0.24). The results demonstrate that evolutionarily optimized ensemble learning provides a reliable and computationally efficient strategy for thermophysical property prediction in nano-PEG composites, offering a practical alternative to extensive experimental characterization.