surrogate models
Surrogate models are approximations of complex models that are used to simulate behavior for optimization or analysis purposes. They provide a simpler and computationally cheaper alternative to assess changes or scenarios in a high-dimensional space.
- CAMO: Convergence-Aware Multi-Fidelity Bayesian Optimization
- Distributionally Robust Performative Optimization
- EngiBench: A Framework for Data-Driven Engineering Design Research
- MAP Estimation with Denoisers: Convergence Rates and Guarantees
- Non-Adaptive Adversarial Face Generation
- Per-Architecture Training-Free Metric Optimization for Neural Architecture Search
- SMRS: advocating a unified reporting standard for surrogate models in the artificial intelligence era.
- TransferBench: Benchmarking Ensemble-based Black-box Transfer Attacks
- Uncertainty-Aware Multi-Objective Reinforcement Learning-Guided Diffusion Models for 3D De Novo Molecular Design