mlsanityData to 7 Oct 2026

Generative Pretraining for Black-Box Optimization

arXiv 2206.10786 ↗First posted 22 Jun 2022cs.LG cs.AIcited by 36

Not on the map. Why?

Abstract (arXiv)

Many problems in science and engineering involve optimizing an expensive black-box function over a high-dimensional space. For such black-box optimization (BBO) problems, we typically assume a small budget for online function evaluations, but also often have access to a fixed, offline dataset for pretraining. Prior approaches seek to utilize the offline data to approximate the function or its inverse but are not sufficiently accurate far from the data distribution. We propose BONET, a generative framework for pretraining a novel black-box optimizer using offline datasets. In BONET, we train an autoregressive model on fixed-length trajectories derived from an offline dataset. We design a sampling strategy to synthesize trajectories from offline data using a simple heuristic of rolling out monotonic transitions from low-fidelity to high-fidelity samples. Empirically, we instantiate BONET using a causally masked Transformer and evaluate it on Design-Bench, where we rank the best on average, outperforming state-of-the-art baselines.

Key facts. Cited by 36 papers in the map (data through 7 Oct 2026). Builds on 33 of its 60 references that are in the map. First posted on arXiv 22 Jun 2022.

Builds on

33 of 60 references.

References · 33

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2020End-to-End Object Detection with Transformerscited by 7,384
  4. 2014Conditional Generative Adversarial Netscited by 2,152
  5. 2016OpenAI Gymcited by 1,684
  6. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,528
  7. 2021Offline Reinforcement Learning as One Big Sequence Modeling Problemcited by 636
  8. 2016Learning to learn by gradient descent by gradient descentcited by 616
  9. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 504
  10. 2018Conditional Neural Processescited by 421
  11. 2018Neural Processescited by 307
  12. 2022Online Decision Transformercited by 194
21 more
  1. 2018Deep Bayesian Bandits Showdown: An Empirical Comparison of Bayesian Deep Networks for Thompson Samplingcited by 166
  2. 2019Convolutional Conditional Neural Processescited by 139
  3. 2020Bayesian Neural Networks: An Introduction and Surveycited by 130
  4. 2019Reinforcement Learning Upside Down: Don't Predict Rewards -- Just Map Them to Actionscited by 124
  5. 2019Conditioning by adaptive sampling for robust designcited by 110
  6. 2022Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimizationcited by 96
  7. 2021Conservative Objective Models for Effective Offline Model-Based Optimizationcited by 87
  8. 2019Model Inversion Networks for Model-Based Optimizationcited by 83
  9. 2019ROBEL: Robotics Benchmarks for Learning with Low-Cost Robotscited by 70
  10. 2017The reparameterization trick for acquisition functionscited by 61
  11. 2019Sequential Neural Processescited by 55
  12. 2020Autofocused oracles for model-based designcited by 50
  13. 2016On Explore-Then-Commit Strategiescited by 46
  14. 2021RoMA: Robust Model Adaptation for Offline Model-based Optimizationcited by 39
  15. 2021Offline Model-Based Optimization via Normalized Maximum Likelihood Estimationcited by 38
  16. 2018Best arm identification in multi-armed bandits with delayed feedbackcited by 29
  17. 2019Knowing The What But Not The Where in Bayesian Optimizationcited by 20
  18. 2021JUMBO: Scalable Multi-task Bayesian Optimization using Offline Datacited by 8
  19. 2021Learning from an Exploring Demonstrator: Optimal Reward Estimation for Banditscited by 8
  20. 2016A sequential Monte Carlo approach to Thompson sampling for Bayesian optimizationcited by 7
  21. 2021Bayesian Neural Networks: Essentialscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 4; 2024: 14; 2025: 9; 2026: 7.

36 papers.

Sort:

By area:

  1. 2023Diffusion Models for Black-Box Optimizationcited by 75
  2. 2022Masked Autoencoding for Scalable and Generalizable Decision Makingcited by 44
  3. 2024LICO: Large Language Models for In-Context Molecular Optimizationcited by 26
  4. 2023ExPT: Synthetic Pretraining for Few-Shot Experimental Designcited by 22
  5. 2024Position: Leverage Foundational Models for Black-Box Optimizationcited by 21
  6. 2024Guided Trajectory Generation with Diffusion Models for Offline Model-based Optimizationcited by 15
  7. 2024Reinforced In-Context Black-Box Optimizationcited by 14
  8. 2025Generative Models in Decision Making: A Surveycited by 14
  9. 2025Offline Model-Based Optimization: Comprehensive Reviewcited by 14
  10. 2024Pretrained Optimization Model for Zero-Shot Black Box Optimizationcited by 13
  11. 2024Offline Model-Based Optimization by Learning to Rankcited by 13
  12. 2025Preference-Guided Diffusion for Multi-Objective Offline Optimizationcited by 12
24 more
  1. 2024Generative Adversarial Model-Based Optimization via Source Critic Regularizationcited by 10
  2. 2024Design Editing for Offline Model-based Optimizationcited by 10
  3. 2023Decision Stacks: Flexible Reinforcement Learning via Modular Generative Modelscited by 8
  4. 2024"Noisier" Noise Contrastive Eestimation is (Almost) Maximum Likelihoodcited by 8
  5. 2024Robust Guided Diffusion for Offline Black-Box Optimizationcited by 8
  6. 2022Reliable Conditioning of Behavioral Cloning for Offline Reinforcement Learningcited by 6
  7. 2025Boosting Offline Optimizers with Surrogate Sensitivitycited by 6
  8. 2025Towards Universal Offline Black-Box Optimization via Learning Language Model Embeddingscited by 6
  9. 2025Incorporating Surrogate Gradient Norm to Improve Offline Optimization Techniquescited by 5
  10. 2025ROOT: Rethinking Offline Optimization as Distributional Translation via Probabilistic Bridgecited by 4
  11. 2024Robust Adaptation of Foundation Models with Black-Box Visual Promptingcited by 3
  12. 2024Cliqueformer: Model-Based Optimization with Structured Transformerscited by 3
  13. 2023ROMO: Retrieval-enhanced Offline Model-based Optimizationcited by 1
  14. 2024From Function to Distribution Modeling: A PAC-Generative Approach to Offline Optimizationcited by 1
  15. 2025Learning Design-Score Manifold to Guide Diffusion Models for Offline Optimizationcited by 1
  16. 2025ZeroShotOpt: Towards Zero-Shot Pretrained Models for Efficient Black-Box Optimizationcited by 1
  17. 2026Diffusion Large Language Models for Black-Box Optimizationcited by 1
  18. 2026Support-Proximity Augmented Diffusion Estimation for Offline Black-Box Optimizationcited by 1
  19. 2026Conformal Candidate Certification for Offline Model-Based Optimizationcited by 1
  20. 2024Offline Stochastic Optimization of Black-Box Objective Functionscited by 0
  21. 2026The Offline-Frontier Shift: Diagnosing Distributional Limits in Generative Multi-Objective Optimizationcited by 0
  22. 2026Black-Box Optimization From Small Offline Datasets via Meta Learning with Synthetic Taskscited by 0
  23. 2026Generative Refinement for Low-Budget Black-Box Optimizationcited by 0
  24. 2026Rethinking Learnability in Offline Data-driven Optimizationcited by 0