mlsanityData to 14 Oct 2026

Bootstrapped Training of Score-Conditioned Generator for Offline Design of Biological Sequences

arXiv 2306.03111 ↗First posted 5 Jun 2023q-bio.QM cs.LG stat.MLcited by 23

Not on the map. Why?

Abstract (arXiv)

We study the problem of optimizing biological sequences, e.g., proteins, DNA, and RNA, to maximize a black-box score function that is only evaluated in an offline dataset. We propose a novel solution, bootstrapped training of score-conditioned generator (BootGen) algorithm. Our algorithm repeats a two-stage process. In the first stage, our algorithm trains the biological sequence generator with rank-based weights to enhance the accuracy of sequence generation based on high scores. The subsequent stage involves bootstrapping, which augments the training dataset with self-generated data labeled by a proxy score function. Our key idea is to align the score-based generation with a proxy score function, which distills the knowledge of the proxy score function to the generator. After training, we aggregate samples from multiple bootstrapped generators and proxies to produce a diverse design. Extensive experiments show that our method outperforms competitive baselines on biological sequential design tasks. We provide reproducible source code: hrefhttps://github.com/kaist-silab/bootgenhttps://github.com/kaist-silab/bootgen.

Key facts. Cited by 23 papers in the map (data through 14 Oct 2026). Builds on 29 of its 56 references that are in the map. First posted on arXiv 5 Jun 2023.

Builds on

29 of 56 references.

References · 29

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  3. 2014Conditional Generative Adversarial Netscited by 2,162
  4. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,546
  5. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,380
  6. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,311
  7. 2021Flow Network based Generative Models for Non-Iterative Diverse Candidate Generationcited by 392
  8. 2022Biological Sequence Design with GFlowNetscited by 174
  9. 2020ProGen: Language Modeling for Protein Generationcited by 128
  10. 2020Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retrainingcited by 123
  11. 2022Equivariant 3D-Conditional Diffusion Models for Molecular Linker Designcited by 122
  12. 2019Conditioning by adaptive sampling for robust designcited by 111
17 more
  1. 2022Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimizationcited by 97
  2. 2021Conservative Objective Models for Effective Offline Model-Based Optimizationcited by 88
  3. 2019Model Inversion Networks for Model-Based Optimizationcited by 83
  4. 2020AdaLead: A simple and robust adaptive greedy search algorithm for sequence designcited by 65
  5. 2017The reparameterization trick for acquisition functionscited by 63
  6. 2020BOSS: Bayesian Optimization over String Spacescited by 54
  7. 2020Autofocused oracles for model-based designcited by 50
  8. 2021RoMA: Robust Model Adaptation for Offline Model-based Optimizationcited by 39
  9. 2021Offline Model-Based Optimization via Normalized Maximum Likelihood Estimationcited by 38
  10. 2022Bootstrapped Transformer for Offline Reinforcement Learningcited by 37
  11. 2022Bidirectional Learning for Offline Infinite-width Model-based Optimizationcited by 35
  12. 2022Structure-aware Protein Self-supervised Learningcited by 33
  13. 2018Design by adaptive samplingcited by 32
  14. 2023Bidirectional Learning for Offline Model-based Biological Sequence Designcited by 20
  15. 2021Deep Extrapolation for Attribute-Enhanced Generationcited by 18
  16. 2022No More Pesky Hyperparameters: Offline Hyperparameter Tuning for RLcited by 6
  17. 2022Consistent Training via Energy-Based GFlowNets for Modeling Discrete Joint Distributionscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 11; 2025: 8; 2026: 3.

23 papers.

Sort:

By area:

  1. 2024Genetic-guided GFlowNets for Sample Efficient Molecular Optimizationcited by 39
  2. 2024Diffusion Models as Constrained Samplers for Optimization with Unknown Constraintscited by 35
  3. 2023Learning to Scale Logits for Temperature-Conditional GFlowNetscited by 27
  4. 2024GTA: Generative Trajectory Augmentation with Guidance for Offline Reinforcement Learningcited by 22
  5. 2024Guided Trajectory Generation with Diffusion Models for Offline Model-based Optimizationcited by 15
  6. 2025Offline Model-Based Optimization: Comprehensive Reviewcited by 15
  7. 2024Offline Multi-Objective Optimizationcited by 14
  8. 2024Robust Optimization in Protein Fitness Landscapes Using Reinforcement Learning in Latent Spacecited by 13
  9. 2024Improved Off-policy Reinforcement Learning in Biological Sequence Designcited by 13
  10. 2024Offline Model-Based Optimization by Learning to Rankcited by 13
  11. 2025Posterior Inference with Diffusion Models for High-dimensional Black-box Optimizationcited by 13
  12. 2024Generative Adversarial Model-Based Optimization via Source Critic Regularizationcited by 10
11 more
  1. 2024Design Editing for Offline Model-based Optimizationcited by 10
  2. 2024"Noisier" Noise Contrastive Eestimation is (Almost) Maximum Likelihoodcited by 8
  3. 2025Towards Universal Offline Black-Box Optimization via Learning Language Model Embeddingscited by 7
  4. 2025Neural Genetic Search in Discrete Spacescited by 5
  5. 2025Posterior Inference in Latent Space for Scalable Constrained Black-box Optimizationcited by 5
  6. 2025ROOT: Rethinking Offline Optimization as Distributional Translation via Probabilistic Bridgecited by 4
  7. 2025Diffusion Alignment as Variational Expectation-Maximizationcited by 4
  8. 2025Overconfident Oracles: Limitations of In Silico Sequence Design Benchmarkingcited by 3
  9. 2026Pareto-Conditioned Diffusion Models for Offline Multi-Objective Optimizationcited by 2
  10. 2026Sample Efficient Generative Molecular Optimization with Joint Self-Improvementcited by 0
  11. 2026Black-Box Optimization From Small Offline Datasets via Meta Learning with Synthetic Taskscited by 0