ABSTRACT
We consider a new class of expensive, resource-constrained optimization problems (here arising from molecular discovery) where costs are associated with the experiments (or evaluations) to be carried out during the optimization process. In the molecular discovery problem, candidate compounds to be optimized must be synthesized in an iterative process that starts from a set of purchasable items and builds up to larger molecules. To produce target molecules, their required resources are either used from already-synthesized items in storage or produced themselves on-demand at an additional cost. Any remaining resources from the production process are stored for reuse for the next evaluations. We model these resource dependencies with a directed acyclic production graph describing the development process from granular purchasable items to evaluable target compounds. Moreover, we develop several resource-eficient algorithms to address this problem. In particular, we develop resource-aware variants of Random Search heuristics and of Bayesian Optimization and analyze their performance in terms of anytime behavior. The experimental results were obtained from a real-world molecular optimization problem. Our results suggest that algorithms that encourage exploitation by reusing existing resources achieve satisfactory results while using fewer resources overall.
- Richard Allmendinger. 2012. Tuning Evolutionary Search for Closed-Loop Optimization. Ph.D. Dissertation. The University of Manchester, UK.Google Scholar
- Richard Allmendinger and Joshua Knowles. 2010. On-Line Purchasing Strategies for an Evolutionary Algorithm Performing Resource-Constrained Optimization. In Parallel Problem Solving from Nature, PPSN XI, Robert Schaefer, Carlos Cotta, Joanna Kołodziej, and Günter Rudolph (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 161--170. Google ScholarCross Ref
- Richard Allmendinger and Joshua D. Knowles. 2011. Policy Learning in Resource-Constrained Optimization. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO 2011, Natalio Krasnogor and Pier Luca Lanzi (Eds.). ACM Press, New York, NY, 1971--1979. Google ScholarDigital Library
- Richard Allmendinger and Joshua D. Knowles. 2013. On Handling Ephemeral Resource Constraints in Evolutionary Search. Evolutionary Computation 21, 3 (Sept. 2013), 497--531. Google ScholarDigital Library
- G. Richard Bickerton, Gaia V. Paolini, Jérémy Besnard, Sorel Muresan, and Andrew L. Hopkins. 2012. Quantifying the Chemical Beauty of Drugs. Nature Chemistry 4, 2 (Feb. 2012), 90--98. Google ScholarCross Ref
- Jürgen Branke and C. Schmidt. 2005. Faster Convergence by Means of Fitness Estimation. Soft Computing 9, 1 (Jan. 2005), 13--20. Google ScholarDigital Library
- M. A. DeRousseau, J. R. Kasprzyk, and W. V. Srubar. 2018. Computational Design Optimization of Concrete Mixtures: A Review. Cement and Concrete Research 109 (July 2018), 42--53. Google ScholarCross Ref
- Douglas H. Fisher. 2016. Recent Advances in AI for Computational Sustainability. IEEE Intelligent Systems 31, 4 (July 2016), 56--61. Google ScholarDigital Library
- Peter I. Frazier. 2018. A Tutorial on Bayesian Optimization. arXiv:1807.02811Google Scholar
- Carla Gomes, Thomas Dietterich, Christopher Barrett, Jon Conrad, Bistra Dilkina, Stefano Ermon, Fei Fang, Andrew Farnsworth, Alan Fern, Xiaoli Fern, Daniel Fink, Douglas Fisher, Alexander Flecker, Daniel Freund, Angela Fuller, John Gregoire, John Hopcroft, Steve Kelling, Zico Kolter, Warren Powell, Nicole Sintov, John Selker, Bart Selman, Daniel Sheldon, David Shmoys, Milind Tambe, Weng-Keen Wong, Christopher Wood, Xiaojian Wu, Yexiang Xue, Amulya Yadav, Abdul-Aziz Yakubu, and Mary Lou Zeeman. 2019. Computational Sustainability: Computing for a Better World and a Sustainable Future. Commun. ACM 62, 9 (Aug. 2019), 56--65. Google ScholarDigital Library
- Carla P. Gomes. 2009. Computational Sustainability: Computational Methods for a Sustainable Environment, Economy, and Society. The Bridge 39, 4 (2009), 5--13.Google Scholar
- Carla P. Gomes, Daniel Fink, R. Bruce van Dover, and John M. Gregoire. 2021. Computational Sustainability Meets Materials Science. Nature Reviews Materials 6, 8 (Aug. 2021), 645--647. Google ScholarCross Ref
- Rafael Gómez-Bombarelli, Jennifer N. Wei, David Duvenaud, José Miguel Hernández-Lobato, Benjamín Sánchez-Lengeling, Dennis Sheberla, Jorge Aguilera-Iparraguirre, Timothy D. Hirzel, Ryan P. Adams, and Alán Aspuru-Guzik. 2018. Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules. ACS Central Science 4, 2 (Feb. 2018), 268--276. arXiv:1610.02415 Google ScholarCross Ref
- Ryan-Rhys Griffiths and José Miguel Hernández-Lobato. 2020. Constrained Bayesian Optimization for Automatic Chemical Design Using Variational Autoencoders. Chemical Science 11, 2 (2020), 577--586. Google ScholarCross Ref
- Nikolaus Hansen, Anne Auger, Raymond Ros, Steffen Finck, and Petr Pošík. 2010. Comparing Results of 31 Algorithms from the Black-Box Optimization Benchmarking BBOB-2009. In Proceedings of the Genetic and Evolutionary Computation Conference Companion, GECCO Companion 2010, Martin Pelikan and Jürgen Branke (Eds.). ACM Press, New York, NY, 1689--1696. Google ScholarDigital Library
- Yaochu Jin. 2011. Surrogate-Assisted Evolutionary Computation: Recent Advances and Future Challenges. Swarm and Evolutionary Computation 1, 2 (June 2011), 61--70. Google Scholar
- Terry Jones and Stephanie Forrest. 1995. Fitness Distance Correlation as a Measure of Problem Difficulty for Genetic Algorithms. In ICGA, Larry J. Eshelman (Ed.). Morgan Kaufmann Publishers, San Francisco, CA, Pittsburgh, PA, 184--192.Google Scholar
- M. C. Kennedy and A. O'Hagan. 2000. Predicting the Output from a Complex Computer Code When Fast Approximations Are Available. Biometrika 87, 1 (March 2000), 1--13. Google ScholarCross Ref
- Ross D. King, Kenneth E. Whelan, Ffion M. Jones, Philip G. K. Reiser, Christopher H. Bryant, Stephen H. Muggleton, Douglas B. Kell, and Stephen G. Oliver. 2004. Functional Genomic Hypothesis Generation and Experimentation by a Robot Scientist. Nature 427, 6971 (Jan. 2004), 247--252. Google ScholarCross Ref
- Joshua D. Knowles. 2009. Closed-loop evolutionary multiobjective optimization. IEEE Computational Intelligence Magazine 4 (2009), 77--91. Issue 3. Google ScholarDigital Library
- Ksenia Korovina, Sailun Xu, Kirthevasan Kandasamy, Willie Neiswanger, Barnabas Poczos, Jeff Schneider, and Eric Xing. 2020. ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations. In Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics. PMLR, Online, 3393--3403.Google Scholar
- Matt J. Kusner, Brooks Paige, and José Miguel Hernández-Lobato. 2017. Grammar Variational Autoencoder. In International Conference on Machine Learning. PMLR, Sydney, Australia, 1945--1954.Google Scholar
- Greg Landrum, Paolo Tosco, Brian Kelley, Ric, Sriniker, Gedeck, Riccardo Vianello, NadineSchneider, Andrew Dalke, Eisuke Kawashima, Dan N, Brian Cole, Matt Swain, Samo Turk, David Cosgrove, AlexanderSavelyev, Alain Vaucher, Gareth Jones, Maciej Wójcikowski, Daniel Probst, Vincent F. Scalfani, Guillaume Godin, Axel Pahl, Francois Berenger, JLVarjo, Strets123, JP, DoliathGavid, Gianluca Sforna, and Jan Holst Jensen. 2021. Rdkit/Rdkit: 2021_03_5 (Q1 2021) Release. Zenodo. Google ScholarCross Ref
- Eric Hans Lee, Valerio Perrone, Cedric Archambeau, and Matthias Seeger. 2020. Cost-Aware Bayesian Optimization. Google ScholarCross Ref
- Manuel López-Ibáñez and Thomas Stützle. 2014. Automatically Improving the Anytime Behaviour of Optimisation Algorithms. European Journal of Operational Research 235, 3 (2014), 569--582. Google ScholarCross Ref
- K. F. Man, K. S. Tang, and S. Kwong. 1996. Genetic Algorithms: Concepts and Applications [in Engineering Design]. IEEE Transactions on Industrial Electronics 43, 5 (Oct. 1996), 519--534. Google ScholarCross Ref
- Harry L. Morgan. 1965. The Generation of a Unique Machine Description for Chemical Structures-a Technique Developed at Chemical Abstracts Service. Journal of chemical documentation 5, 2 (1965), 107--113.Google ScholarCross Ref
- Carl Edward Rasmussen and Christopher K. I. Williams. 2006. Gaussian Processes for Machine Learning. MIT Press, Cambridge, MA.Google ScholarDigital Library
- David Rogers and Mathew Hahn. 2010. Extended-Connectivity Fingerprints. Journal of Chemical Information and Modeling 50, 5 (May 2010), 742--754. Google ScholarCross Ref
- Philip A. Romero, Andreas Krause, and Frances H. Arnold. 2013. Navigating the Protein Fitness Landscape with Gaussian Processes. Proceedings of the National Academy of Sciences 110, 3 (2013), E193--E201.Google ScholarCross Ref
- Philippe Schwaller, Théophile Gaudin, Dávid Lányi, Costas Bekas, and Teodoro Laino. 2018. "Found in Translation": Predicting Outcomes of Complex Organic Chemistry Reactions Using Neural Sequence-to-Sequence Models. Chemical Science 9, 28 (2018), 6091--6098. Google ScholarCross Ref
- Jasper Snoek, Hugo Larochelle, and Ryan P. Adams. 2012. Practical Bayesian Optimization of Machine Learning Algorithms. In 26th Annual Conference on Neural Information Processing Systems (Advances in Neural Information Processing Systems, Vol. 4). Morgan Kaufmann Publishers, Inc., Lake Tahoe, USA, 2951--2959.Google Scholar
- Thomas Spangenberg, Jeremy N. Burrows, Paul Kowalczyk, Simon McDonald, Timothy N. C. Wells, and Paul Willis. 2013. The Open Access Malaria Box: A Drug Discovery Catalyst for Neglected Diseases. PLOS ONE 8, 6 (June 2013), e62906. Google ScholarCross Ref
- Kei Terayama, Masato Sumita, Ryo Tamura, and Koji Tsuda. 2021. Black-Box Optimization for Automated Discovery. Accounts of Chemical Research 54, 6 (March 2021), 1334--1346. Google ScholarCross Ref
- David Weininger. 1988. SMILES, a Chemical Language and Information System. 1. Introduction to Methodology and Encoding Rules. Journal of chemical information and computer sciences 28, 1 (1988), 31--36.Google ScholarDigital Library
- Dawei Zhan and Huanlai Xing. 2020. Expected Improvement for Expensive Optimization: A Review. Journal of Global Optimization 78, 3 (Nov. 2020), 507--544. Google ScholarDigital Library
- Shlomo Zilberstein. 1996. Using Anytime Algorithms in Intelligent Systems. AI Magazine 17, 3 (1996), 73--83. Google ScholarDigital Library
Index Terms
- Expensive optimization with production-graph resource constraints: a first look at a new problem class
Recommendations
A Pre-initialization Stage of Population-Based Bio-inspired Metaheuristics for Handling Expensive Optimization Problems
ADMA 2013: Part II of the Proceedings of the 9th International Conference on Advanced Data Mining and Applications - Volume 8347Metaheuristics are probabilistic optimization algorithms which are applicable to a wide range of optimization problems. Bio-inspired, also called nature-inspired, optimization algorithms are the most widely-known metaheuristics. The general scheme of ...
A multi-strategy surrogate-assisted competitive swarm optimizer for expensive optimization problems
AbstractEvolutionary computation is a powerful tool for solving nonconvex optimization problems. Generally, evolutionary algorithms take numerous fitness evaluations to obtain the potential optimal solutions. This poses a critical challenge for applying ...
Highlights- SACSO is proposed to solve expensive optimization problems and applied to speed reducer design.
- Particle selection criteria contain global search, local search, and opposition-based search.
- A dynamic adaptation strategy is proposed ...
A survey on evolutionary computation for complex continuous optimization
AbstractComplex continuous optimization problems widely exist nowadays due to the fast development of the economy and society. Moreover, the technologies like Internet of things, cloud computing, and big data also make optimization problems with more ...
Comments