Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
102 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Enhanced Bayesian Optimization via Preferential Modeling of Abstract Properties (2402.17343v1)

Published 27 Feb 2024 in cs.LG and stat.ML

Abstract: Experimental (design) optimization is a key driver in designing and discovering new products and processes. Bayesian Optimization (BO) is an effective tool for optimizing expensive and black-box experimental design processes. While Bayesian optimization is a principled data-driven approach to experimental optimization, it learns everything from scratch and could greatly benefit from the expertise of its human (domain) experts who often reason about systems at different abstraction levels using physical properties that are not necessarily directly measured (or measurable). In this paper, we propose a human-AI collaborative Bayesian framework to incorporate expert preferences about unmeasured abstract properties into the surrogate modeling to further boost the performance of BO. We provide an efficient strategy that can also handle any incorrect/misleading expert bias in preferential judgments. We discuss the convergence behavior of our proposed framework. Our experimental results involving synthetic functions and real-world datasets show the superiority of our method against the baselines.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (43)
  1. Ruben Martinez-Cantin. BayesOpt: a Bayesian optimization library for nonlinear optimization, experimental design and bandits. J. Mach. Learn. Res., 15(1):3735–3739, 2014.
  2. Bayesian optimization for adaptive experimental design: a review. IEEE access, 8:13937–13948, 2020.
  3. Information-theoretic regret bounds for Gaussian process optimization in the bandit setting. IEEE Transactions on Information Theory, 58(5):3250–3265, 2012.
  4. On kernelized multi-armed bandits. In International Conference on Machine Learning, pages 844–853. PMLR, 2017a.
  5. Kevin Swersky. Improving Bayesian optimization for machine learning using expert priors. University of Toronto (Canada), 2017.
  6. Bayesian optimization for objective functions with varying smoothness. In Australasian Joint Conference on Artificial Intelligence, pages 460–472, 2019.
  7. Accelerating experimental design by incorporating experimenter hunches. In 2018 IEEE International Conference on Data Mining (ICDM), pages 257–266, 2018. doi:10.1109/ICDM.2018.00041.
  8. π𝜋\piitalic_πBO: Augmenting Acquisition Functions with User Beliefs for Bayesian Optimization. arXiv preprint arXiv:2204.11051, 2022.
  9. Human-AI Collaborative Bayesian optimization. In Advances in Neural Information Processing Systems, 2022.
  10. Top-k ranking Bayesian optimization. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 9135–9143, 2021.
  11. Gaussian processes for machine learning, volume 2. MIT press Cambridge, MA, 2006.
  12. A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning. arXiv preprint arXiv:1012.2599, 2010.
  13. Peter I Frazier. A tutorial on Bayesian optimization. arXiv preprint arXiv:1807.02811, 2018.
  14. H. J. Kushner. A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise. Journal of Basic Engineering, 86(1):97–106, 03 1964. ISSN 0021-9223. doi:10.1115/1.3653121.
  15. Student-t processes as alternatives to Gaussian processes. In Artificial intelligence and statistics, pages 877–885, 2014.
  16. Degradation data analysis and remaining useful life estimation: A review on Wiener-process-based methods. European Journal of Operational Research, 271(3):775–796, 2018.
  17. The application of Bayesian methods for seeking the extremum. In Towards Global Optimization, volume 2, pages 117–129. September 1978. ISBN 0-444-85171-2.
  18. William R Thompson. On the likelihood that one unknown probability exceeds another in view of the evidence of two samples. Biometrika, 25(3-4):285–294, 1933.
  19. Prospect theory: An analysis of decision under risk. In Handbook of the fundamentals of financial decision making: Part I, pages 99–127. World Scientific, 2013.
  20. A/B testing: The most powerful way to turn clicks into customers. John Wiley & Sons, 2015.
  21. The adaptive web: methods and strategies of web personalization, volume 4321. Springer Science & Business Media, 2007.
  22. TrueSkill: a Bayesian skill rating system. Advances in neural information processing systems, 19, 2006.
  23. Preference learning with Gaussian processes. In Proceedings of the 22nd international conference on Machine learning, pages 137–144, 2005.
  24. Preferential bayesian optimization. In International Conference on Machine Learning, pages 1282–1291. PMLR, 2017.
  25. Projective preferential bayesian optimization. In International Conference on Machine Learning, pages 6884–6892. PMLR, 2020.
  26. Preferential Bayesian optimization with skew Gaussian processes. In Proceedings of the Genetic and Evolutionary Computation Conference Companion, pages 1842–1850, 2021.
  27. Multi-attribute Bayesian optimization with interactive preference learning. In International Conference on Artificial Intelligence and Statistics, pages 4496–4507. PMLR, 2020.
  28. Louis L Thurstone. A law of comparative judgment. In Scaling, pages 81–92. Routledge, 2017.
  29. Learning to optimize via posterior sampling. Mathematics of Operations Research, 39(4):1221–1243, 2014.
  30. Parallelised bayesian optimisation via thompson sampling. In International Conference on Artificial Intelligence and Statistics, pages 133–142, 2018.
  31. On kernelized multi-armed bandits. In Doina Precup and Yee Whye Teh, editors, Proceedings of the 34th International Conference on Machine Learning, volume 70 of Proceedings of Machine Learning Research, pages 844–853, International Convention Centre, Sydney, Australia, Aug 2017b. PMLR.
  32. Understanding the eluder dimension. In Advances in Neural Information Processing Systems, 2021.
  33. Radford M Neal. Bayesian learning for neural networks, volume 118. Springer Science & Business Media, 2012.
  34. S. Surjanovic and D. Bingham. Virtual library of simulation experiments: Test Functions and Datasets, 2017. URL "http://www.sfu.ca/~ssurjano". [Online; accessed 21-January-2023].
  35. Data-driven assessment of electrode calendering process by combining experimental results, in silico mesostructures generation and machine learning. Journal of Power Sources, 480:229103, 2020.
  36. Formulation and manufacturing optimization of Lithium-ion graphite-based electrodes via machine learning. Cell Reports Physical Science, 2(12):100683, 2021.
  37. Nachman Aronszajn. Theory of reproducing kernels. Transactions of the American mathematical society, 68(3):337–404, 1950.
  38. Thompson sampling: An asymptotically optimal finite-time analysis. In Algorithmic Learning Theory: 23rd International Conference, ALT 2012, Lyon, France, October 29-31, 2012. Proceedings 23, pages 199–213. Springer, 2012.
  39. An entropy search portfolio for Bayesian optimization. arXiv preprint arXiv:1406.4625, 2014.
  40. A sequential Monte Carlo approach to Thompson sampling for Bayesian optimization. arXiv preprint arXiv:1604.00169, 2016.
  41. The future of Lithium and Lithium-ion batteries in implantable medical devices. Journal of power sources, 97:742–746, 2001.
  42. Lithium-ion (Li-ion) battery technology evolves to serve an extended range of telecom applications. In 2011 IEEE 33rd International Telecommunications Energy Conference (INTELEC), pages 1–9. IEEE, 2011.
  43. Recycling Lithium-ion batteries from electric vehicles. nature, 575(7781):75–86, 2019.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (6)
  1. Arun Kumar A V (3 papers)
  2. Alistair Shilton (14 papers)
  3. Sunil Gupta (78 papers)
  4. Santu Rana (68 papers)
  5. Stewart Greenhill (2 papers)
  6. Svetha Venkatesh (160 papers)
Citations (3)
View on Semantic Scholar
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets