Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
November 20, 2025Structural and Multidisciplinary OptimizationOpen Access

Scalable neural network-based blackbox optimization

View Full Paper
Ask AI
Bookmark
Share

Authors

PKPavankumar KoratikereLLLeifur Leifsson

Discussion

Loading...

Member takes

Overview

Proposed scalable neural network-based optimization improves exploration and sampling, reducing evaluations by 40-60% in high dimensions.

Key Points

  • SNBO significantly improves function values compared to the best baseline algorithm across many test problems and dimensions.
  • Utilizing advanced neural networks, the new method handles blackbox optimization effectively while avoiding complex uncertainty estimates.
  • Incorporating both exploration and exploitation, the scalable approach adapts its sampling region for better performance in challenging optimization tasks.
  • The utilization of the Friedman test establishes SNBO's improved statistical performance compared to traditional Bayesian Optimization methods.

Cite This Study

Koratikere et al. (2025) studied this question.

synapsesocial.com/papers/6924f074c0ce034ddc34fa7dhttps://doi.org/10.1007/s00158-025-04195-5
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1When to Explore and When to Exploit: Adaptive Decisions in Bayesian Optimization2026
  2. 2Machine Learning Algorithms for Improving Black Box Optimization Solvers2025
  3. 3NM2-BO: Non-Myopic Multifidelity Bayesian Optimization2024 · 6 citations
  4. 4Comparison of High-Dimensional Bayesian Optimization Algorithms on BBOB2024 · 28 citations
  5. 5Quantile-Scaled Bayesian Optimization Using Rank-Only Feedback2025