mlsanityData to 14 Oct 2026

Boosting Offline Optimizers with Surrogate Sensitivity

arXiv 2503.04181 ↗First posted 6 Mar 2025cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Offline optimization is an important task in numerous material engineering domains where online experimentation to collect data is too expensive and needs to be replaced by an in silico maximization of a surrogate of the black-box function. Although such a surrogate can be learned from offline data, its prediction might not be reliable outside the offline data regime, which happens when the surrogate has narrow prediction margin and is (therefore) sensitive to small perturbations of its parameterization. This raises the following questions: (1) how to regulate the sensitivity of a surrogate model; and (2) whether conditioning an offline optimizer with such less sensitive surrogate will lead to better optimization performance. To address these questions, we develop an optimizable sensitivity measurement for the surrogate model, which then inspires a sensitivity-informed regularizer that is applicable to a wide range of offline optimizers. This development is both orthogonal and synergistic to prior research on offline optimization, which is demonstrated in our extensive experiment benchmark.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 19 of its 28 references that are in the map. First posted on arXiv 6 Mar 2025.

Builds on

19 of 28 references.

References · 19

  1. 2014Conditional Generative Adversarial Netscited by 2,162
  2. 2016OpenAI Gymcited by 1,736
  3. 2019Conditioning by adaptive sampling for robust designcited by 111
  4. 2022Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimizationcited by 97
  5. 2021Conservative Objective Models for Effective Offline Model-Based Optimizationcited by 88
  6. 2019Model Inversion Networks for Model-Based Optimizationcited by 83
  7. 2023Diffusion Models for Black-Box Optimizationcited by 76
  8. 2019ROBEL: Robotics Benchmarks for Learning with Low-Cost Robotscited by 72
  9. 2017Stochastic Zeroth-order Optimization in High Dimensionscited by 63
  10. 2020Autofocused oracles for model-based designcited by 50
  11. 2021Local Stochastic Gradient Descent Ascent: Convergence Analysis and Communication Efficiencycited by 40
  12. 2021RoMA: Robust Model Adaptation for Offline Model-based Optimizationcited by 39
7 more
  1. 2021Offline Model-Based Optimization via Normalized Maximum Likelihood Estimationcited by 38
  2. 2022Generative Pretraining for Black-Box Optimizationcited by 36
  3. 2023Importance-aware Co-teaching for Offline Model-based Optimizationcited by 28
  4. 2023Parallel-mentoring for Offline Model-based Optimizationcited by 26
  5. 2023ExPT: Synthetic Pretraining for Few-Shot Experimental Designcited by 22
  6. 2024Offline Model-Based Optimization via Policy-Guided Gradient Searchcited by 19
  7. 2021Measuring the robustness of Gaussian processes to kernel choicecited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 2.

6 papers.

  1. 2025Offline Model-Based Optimization: Comprehensive Reviewcited by 15
  2. 2025Towards Universal Offline Black-Box Optimization via Learning Language Model Embeddingscited by 7
  3. 2025ROOT: Rethinking Offline Optimization as Distributional Translation via Probabilistic Bridgecited by 4
  4. 2026On the Learnability of Offline Model-Based Optimization: A Ranking Perspectivecited by 2
  5. 2025Learning Design-Score Manifold to Guide Diffusion Models for Offline Optimizationcited by 1
  6. 2026Black-Box Optimization From Small Offline Datasets via Meta Learning with Synthetic Taskscited by 0