Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
60 tokens/sec
GPT-4o
12 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Goal-conditioned GFlowNets for Controllable Multi-Objective Molecular Design (2306.04620v2)

Published 7 Jun 2023 in cs.LG and q-bio.BM

Abstract: In recent years, in-silico molecular design has received much attention from the machine learning community. When designing a new compound for pharmaceutical applications, there are usually multiple properties of such molecules that need to be optimised: binding energy to the target, synthesizability, toxicity, EC50, and so on. While previous approaches have employed a scalarization scheme to turn the multi-objective problem into a preference-conditioned single objective, it has been established that this kind of reduction may produce solutions that tend to slide towards the extreme points of the objective space when presented with a problem that exhibits a concave Pareto front. In this work we experiment with an alternative formulation of goal-conditioned molecular generation to obtain a more controllable conditional model that can uniformly explore solutions along the entire Pareto front.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. Concrete problems in ai safety. arXiv preprint arXiv:1606.06565, 2016.
  2. Hindsight experience replay. Advances in neural information processing systems, 30, 2017.
  3. Dyngfn: Bayesian dynamic causal discovery using generative flow networks. arXiv preprint arXiv:2302.04178, 2023.
  4. Performance indicators in multiobjective optimization. European journal of operational research, 292(2):397–422, 2021.
  5. Flow network based generative models for non-iterative diverse candidate generation. Advances in Neural Information Processing Systems, 34:27381–27394, 2021a.
  6. Gflownet foundations. arXiv preprint arXiv:2111.09266, 2021b.
  7. Quantifying the chemical beauty of drugs. Nature chemistry, 4(2):90–98, 2012.
  8. A novel workflow for the inverse qspr problem using multiobjective optimization. Journal of computer-aided molecular design, 20:333–341, 2006.
  9. Guacamol: benchmarking models for de novo molecular design. Journal of chemical information and modeling, 59(3):1096–1108, 2019.
  10. Solving multiobjective optimization problems using an artificial immune system. Genetic programming and evolvable machines, 6:163–190, 2005.
  11. Bayesian structure learning with generative flow networks. In Uncertainty in Artificial Intelligence, pp.  518–528. PMLR, 2022.
  12. Ehrgott, M. Multicriteria optimization, volume 491. Springer Science & Business Media, 2005.
  13. A tutorial on multiobjective optimization: fundamentals and evolutionary methods. Natural computing, 17:585–609, 2018.
  14. Estimation of synthetic accessibility score of drug-like molecules based on molecular complexity and fragment contributions. Journal of cheminformatics, 1:1–11, 2009.
  15. Addressing function approximation error in actor-critic methods. In International conference on machine learning, pp. 1587–1596. PMLR, 2018.
  16. Gflownet-em for learning compositional latent variable models. arXiv preprint arXiv:2302.06576, 2023.
  17. Therapeutics data commons: Machine learning datasets and tasks for drug discovery and development. arXiv preprint arXiv:2102.09548, 2021.
  18. Biological sequence design with gflownets. In International Conference on Machine Learning, pp. 9786–9801. PMLR, 2022a.
  19. Multi-objective gflownets. arXiv preprint arXiv:2210.12765, 2022b.
  20. Multi-objective molecule generation using interpretable substructures. In International conference on machine learning, pp. 4849–4859. PMLR, 2020.
  21. Decisions with Multiple Objectives: Preferences and Value Trade-Offs. Wiley series in probability and mathematical statistics. Applied probability and statistics. Cambridge University Press, 1993. ISBN 9780521438834. URL https://books.google.ca/books?id=GPE6ZAqGrnoC.
  22. Fragment based drug design: from experimental to computational approaches. Current medicinal chemistry, 19(30):5128–5147, 2012.
  23. A theory of continuous generative flow networks. arXiv preprint arXiv:2301.12594, 2023.
  24. Pareto multi-task learning. Advances in neural information processing systems, 32, 2019.
  25. Learning gflownets from partial episodes for improved convergence and stability. arXiv preprint arXiv:2209.12782, 2022.
  26. Trajectory balance: Improved credit assignment in gflownets. arXiv preprint arXiv:2201.13259, 2022a.
  27. Gflownets and variational inference. arXiv preprint arXiv:2210.00580, 2022b.
  28. Miettinen, K. Nonlinear multiobjective optimization, volume 12. Springer Science & Business Media, 2012.
  29. Human-level control through deep reinforcement learning. nature, 518(7540):529–533, 2015.
  30. Better training of gflownets with local credit and incomplete trajectories. arXiv preprint arXiv:2302.01687, 2023.
  31. Multiobjective molecular design for integrated process-solvent systems synthesis. AIChE Journal, 52(3):1057–1070, 2006.
  32. Non-convex multi-objective optimization. Springer, 2017.
  33. Recipe for a general, powerful, scalable graph transformer. Advances in Neural Information Processing Systems, 35:14501–14515, 2022.
  34. Direct behavior specification via constrained reinforcement learning. arXiv preprint arXiv:2112.12228, 2021.
  35. Universal value function approximators. In International conference on machine learning, pp. 1312–1320. PMLR, 2015.
  36. Deep reinforcement learning for multiparameter optimization in de novo drug design. Journal of chemical information and modeling, 59(7):3166–3176, 2019.
  37. Moleculenet: a benchmark for molecular machine learning. Chemical science, 9(2):513–530, 2018.
  38. Graph transformer networks. Advances in neural information processing systems, 32, 2019.
  39. Unifying generative models with gflownets. arXiv preprint arXiv:2209.02606, 2022.
  40. Robust scheduling with gflownets. arXiv preprint arXiv:2302.05446, 2023.
  41. Optimization of molecules via deep reinforcement learning. Scientific reports, 9(1):1–10, 2019.
Citations (11)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now