Artificial intelligence techniques from the areas of machine learning and optimization play a crucial role in various applications. Both areas are highly complementary as machine learning models are optimized using optimization techniques and optimization techniques can be made more efficient by incorporating machine learning into these techniques.

In both areas, multitasking approaches have gained significant interest as they allow to solve not just a single given problem but multiple problems with one machine learning or optimization approach. In the area of machine learning, multitasking usually refers to machine learning models that can be used for different tasks [1, 5] whereas ultitasking in the optimization context refers to producing optimized solutions for multiple tasks/problems using a single optimization rum. Approaches of multi-tasking for optimization have gained significant attention in the area of evolutionary algorithms under the term evolutionary multi-tasking [2] as these methods problem great flexibility in terms of solving different problems with a single algorithm run.

Although multi-tasking in the areas of machine learning and optimization shares important aspects and both areas are highly complementary to each other [3, 4], there is insufficient interactions between these two areas. Bringing together researchers from these two fields would booster would be highly beneficial to harvest the immens potential of such interactions.

The goal of this Shonan meeting is to bring together researchers from the areas of machine learning and optimization working on multitask approaches with the goal to establish closer collaborations between the two areas. It will enable us to leverage these complementary techniques and boost multitasking in machine learning and optimization. It is envisioned that this would also allow the creation of more complex artificial intelligence tools that are able to multi-task based on underlying machine learning and optimization techniques.

The primary aim of this proposed Shonan meeting is to provide a collaborative forum for machine learning and optimization in the context of multitasking. It is envisioned that there will be 50% of the participants from each of the two areas covering theoretical foundations as well as high impact applications. The goal of the meeting is to explore and debate this very promising area within machine learning and optimization and provide novel ideas that are highly beneficial to the design of artificial intelligence approaches.

The meeting will include presentations by the participants as well as discussion groups for hot topics and future work. The discussion groups play a central role and should reflect on the current state of the art in the different areas of machine learning and optimization, foster collaborative work, and establish new research directions. The outcomes of these discussion groups will be summarized in the Shonan meeting report such that it is available to all researchers interested in computational intelligence and software engineering.

[1] R. Caruana. Multitask learning. Mach. Learn., 28(1):41–75, 1997.

[2] L. Feng, A. Gupta, K. Tan, and Y. Ong. Evolutionary Multi-Task Optimization: Foundations and Methodologies. Machine Learning: Foundations, Methodologies, and Applications. Springer Nature Singapore, 2023.

[3] O. Sener and V. Koltun. Multi-task learning as multi-objective optimization. In NeurIPS, pages 525–536, 2018.

[4] K. Swersky, J. Snoek, and R. P. Adams. Multi-task bayesian optimization. In NIPS, pages 2004–2012, 2013.

[5] J. Yu, Y. Dai, X. Liu, J. Huang, Y. Shen, K. Zhang, R. Zhou, E. Adhikarla, Ye, Y. Liu, Z. Kong, K. Zhang, Y. Yin, V. Namboodiri, B. D. Davison, H. Moore, and Y. Chen. Unleashing the power of multi-task learning: A comprehensive survey spanning traditional, deep, and pretrained foundation model eras. CoRR, abs/2404.18961, 2024.