Transfer, Multitask & Multiobjective Optimization, Bayesian Optimization — Selected Works

Overview

This research asks how an optimizer can improve with experience instead of restarting every search from zero. Knowledge from earlier or concurrently solved tasks is reused to reduce evaluation cost, accelerate convergence, and improve solution quality. The programme connects evolutionary computation with transfer learning, multiobjective search, surrogate modelling, Gaussian processes, and Bayesian optimization.

Sequential transfer optimization Evolutionary multitasking Multiform optimization Multiobjective optimization Bayesian optimization Gaussian-process surrogates

Research Themes

• Sequential transfer optimization: reuses data, solution representations, or surrogate models from previously solved source tasks to speed up a new target problem, while learning to suppress negative transfer.
• Multitask and multifactorial optimization: solves several tasks together so that useful genetic material, models, or learned representations can move between related searches as they evolve.
• Multiobjective and multiform optimization: searches for trade-off solutions and can solve alternative formulations of a problem together, exploiting shared structure across decision and objective spaces.
• Bayesian and surrogate-assisted optimization: uses probabilistic models—especially Gaussian processes—to guide expensive black-box searches, quantify uncertainty, and transfer information across heterogeneous tasks.

Keynotes

• Ong, Y. S., Foundations of Transfer and Multitask Optimization and Advances with Generative AI and Large Language Models (Seminar / Talk, Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University, 22 July 2024). Details.
• Ong, Y. S., Multi-Task Multi-Objective Optimization: Neuroevolution of Set of Machine Learning Model Sets from Large Models (Keynote, AI Forum 2023, TaiWan, 5 May 2023). Details.
• Ong, Y. S., Generalizable Optimization Intelligence for the Cloud and Edge (Keynote, 9th EAI International Conference on Cloud Computing, CloudComp 2019, Sydney, Australia, 4–5 December 2019). Details.
• Ong, Y. S., “Memes” as Building Blocks of Intelligence: From Refinement to Higher Order Representations in Transfer and Multitask Optimization (Plenary, IEEE World Congress on Computational Intelligence, WCCI 2018, Rio de Janeiro, Brazil, 8–13 July 2018). Details.

Selected Publications

• 2025
  J. Liu, A. Gupta, C. C. Ooi, and Y. S. Ong.
  ExTrEMO: Transfer Evolutionary Multiobjective Optimization With Proof of Faster Convergence, IEEE Transactions on Evolutionary Computation(TEVC 2023), 29(1), 102–116.
• 2024
  Z. Chen, A. Gupta, L. Zhou, and Y. S. Ong.
  Scaling Multiobjective Evolution to Large Data With Minions: A Bayes-Informed Multitask Approach, IEEE Transactions on Cybernetics(TCYB 2022), 54(2), 1294–1307.
• 2024
  J. Liu, A. Gupta, and Y. S. Ong.
  Bayesian Inverse Transfer in Evolutionary Multiobjective Optimization, ACM Transactions on Evolutionary Learning and Optimization, 4(4), 1–27.
• 2023
  C. S. Tan, A. Gupta, Y. S. Ong, M. Pratama, P. S. Tan, and S. K. Lam.
  Pareto Optimization With Small Data by Learning Across Common Objective Spaces, Scientific Reports, 13, 7842.
• 2023
  P. Wei, Y. Ke, Y. S. Ong, and Z. Ma.
  Adaptive Transfer Kernel Learning for Transfer Gaussian Process Regression, IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(6), 7142–7156.
• 2022
  A. Gupta, L. Zhou, Y. S. Ong, Z. Chen, and Y. Hou.
  Half a Dozen Real-World Applications of Evolutionary Multitasking, and More, IEEE Computational Intelligence Magazine, 17(2), 49–66.
• 2022
  L. Bai, W. Lin, A. Gupta, and Y. S. Ong.
  From Multitask Gradient Descent to Gradient-Free Evolutionary Multitasking: A Proof of Faster Convergence, IEEE Transactions on Cybernetics(TCYB 2021), 52(8), 8561–8573.
• 2021
  A. T. W. Min, A. Gupta, and Y. S. Ong.
  Generalizing Transfer Bayesian Optimization to Source-Target Heterogeneity, IEEE Transactions on Automation Science and Engineering, 18(4), 1754–1765.
• 2021
  Z. Guo, Y. S. Ong, and H. Liu.
  Calibrated and Recalibrated Expected Improvements for Bayesian Optimization, Structural and Multidisciplinary Optimization, 64, 3549–3567.
• 2020
  K. K. Bali, Y. S. Ong, A. Gupta, and P. S. Tan.
  Multifactorial Evolutionary Algorithm With Online Transfer Parameter Estimation: MFEA-II, IEEE Transactions on Evolutionary Computation(TEVC 2019), 24(1), 69–83.
• 2019
  B. Da, A. Gupta, and Y. S. Ong.
  Curbing Negative Influences Online for Seamless Transfer Evolutionary Optimization, IEEE Transactions on Cybernetics(TCYB 2018), 49(12), 4365–4378.
• 2019
  A. T. W. Min, Y. S. Ong, A. Gupta, and C. K. Goh.
  Multi-Problem Surrogates: Transfer Evolutionary Multiobjective Optimization of Computationally Expensive Problems, IEEE Transactions on Evolutionary Computation(TEVC 2017), 23(1), 15–28.
• 2018
  A. Gupta, Y. S. Ong, and L. Feng.
  Insights on Transfer Optimization: Because Experience is the Best Teacher, IEEE Transactions on Emerging Topics in Computational Intelligence, 2(1), 51–64.
• 2017
  A. Gupta, Y. S. Ong, L. Feng, and K. C. Tan.
  Multi-Objective Multifactorial Optimization in Evolutionary Multitasking, IEEE Transactions on Cybernetics(TCYB 2016), 47(7), 1652–1665.
• 2016
  A. Gupta, Y. S. Ong, and L. Feng.
  Multifactorial Evolution: Toward Evolutionary Multitasking, IEEE Transactions on Evolutionary Computation(TEVC 2015), 20(3), 343–357.

Books & Resources

• A. Gupta, Y. S. Ong. Memetic Computation: The Mainspring of Knowledge Transfer in a Data-Driven Optimization Era (Springer, 2019).
• L. Feng, A. Gupta, K. C. Tan, Y. S. Ong. Multi-Task Optimization: Foundations and Methodologies (Springer, 2023).
• Transfer Evolutionary Optimization benchmark problems and code.