PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 10, 2026Applied Sciences0 citationsOpen Access

Beyond Accuracy in AI: A Multi-Objective Benchmark of Inductive Bias, Robustness, Computational Efficiency, and Pareto-Optimal Trade-Off

View Full Paper
HOHüseyin Enes OkutanMBMuhammet Baykara

Key Points

  • This study aims to benchmark various classifiers on nonlinear classification tasks to evaluate their performance comprehensively.
  • Conducted 640 controlled experiments with multiple sample sizes and random seeds across four XOR variants.
  • Evaluated classifiers using a multi-objective framework focusing on accuracy, computational efficiency, and resource usage.
  • Analyzed trade-offs using Pareto frontier analysis to identify optimal solutions.
  • MLP_deep achieved the highest overall accuracy across all classifiers.
  • MLP_small demonstrated competitive performance with significantly lower computational costs.
  • High-dimensional XOR scenarios led to reduced performance across all models, indicating increased complexity.

Abstract

Nonlinear classification problems such as XOR are widely used to evaluate machine learning models beyond linear separability. In this study, a comprehensive benchmark is proposed to analyze eight classifiers (Logistic Regression, Linear SVM, RBF SVM, Decision Tree, Random Forest, KNN, MLPₛmall, MLPdeep) across four XOR variants (clean, noisy, rotated, high-dimensional). A total of 640 controlled experiments are conducted using multiple sample sizes and random seeds. Models are evaluated using a multi-objective framework including accuracy, training and inference time, memory usage, energy consumption, and model size. Results show that MLPdeep achieves the highest overall accuracy, while MLPₛmall provides competitive performance with significantly lower computational cost. Decision Tree offers a strong balance between efficiency and accuracy, whereas Random Forest achieves competitive accuracy at higher resource usage. High-dimensional XOR is the most challenging scenario, significantly reducing overall performance across models. Pareto frontier analysis further highlights optimal trade-offs between predictive performance and resource efficiency. The study demonstrates that no single model is universally optimal and emphasizes the importance of resource-aware model selection in nonlinear classification tasks.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Okutan et al. (2026) studied this question.

synapsesocial.com/papers/6a0021e6c8f74e3340f9ce89https://doi.org/10.3390/app16104637
Ask AI
Helpful
Bookmark
Share
View Full Paper