Summary
Optimizing many related tasks at once often struggles when only a few test results are available, causing mistakes from assuming tasks are too connected or not enough. The authors offer a new way called In-Context Guidance Multitask Optimization (ICG-MTO) that uses a powerful pre-learned model to guess how tasks relate to each other without needing lots of examples. This helps guide the optimization better at the start, reducing errors, and gradually lets the real task data take over as more results come in. They tested their method on computer simulations and a robot control problem, showing it works well and can be applied in different optimization algorithms.
What this means in practice
- •For robot control engineers: Improve tuning of robot control parameters across multiple tasks when only limited test runs are possible.
- •For machine learning developers: Enhance multitask hyperparameter search by guiding early optimization stages with pretrained model signals.
Authors
Tingyang Wei, Haofeng Wu, Jiao Liu, Zhao Wei, Puay Siew Tan, Yew-Soon Ong
Abstract
Multi-task optimization (MTO) addresses a set of optimization tasks simultaneously, often suffering from inaccurate inter-task relationship estimation under limited evaluation budgets, leading to negative transfer. This paper introduces In-Context Guidance Multitask Optimization (ICG-MTO), a novel framework that leverages numerical foundational models to improve inter-task coupling estimation in few-shot scenarios. Unlike conventional methods that rely solely on scarce observed data, ICG-MTO employs a frozen foundational model to infer auxiliary guidance through in-context learning. The framework operates through three stages: constructing an algorithm-specific in-context query from evaluated solutions, using the foundational model to infer a guidance signal characterizing predictive relationships among tasks, and translating this signal into algorithm-specific guidance for maximum-a-posteriori coupling estimation. This approach provides regularization during the early, data-scarce stages of optimization and gradually relinquishes control as task-specific observations accumulate. We instantiate the framework in multitask Bayesian optimization as ICG-MTBO, using directional fitness-class queries to guide inter-task coupling estimation, and further instantiate it in MFEA-II using decision-space-overlap queries to guide random mating probability estimation. Experiments across synthetic benchmarks and a real-world robot arm control problem, together with evaluations under different acquisition functions and evolutionary multitasking, demonstrate the effectiveness and generality of ICG-MTO for few-shot multitask optimization.