Continual Human-in-the-Loop Optimization

2025-03-07

AI Summary

To address the inefficiency and difficulty in dynamically adapting cross-user input parameters in human-computer interaction—particularly the challenge of balancing population-level commonalities with individual-specific variations—this paper proposes a continual human-in-the-loop optimization framework. It is the first to integrate the *continual learning* paradigm into human-in-the-loop optimization, employing Bayesian neural networks to jointly model shared user features and personalized biases, and introducing a generative replay mechanism to mitigate catastrophic forgetting. Evaluated on VR touch-typing text entry, the framework leverages historical user data to construct transferable surrogate models, substantially reducing online tuning time for new users. As the user population grows, the average optimization rounds per user decrease by up to 42%. This work establishes a novel, efficient, and scalable paradigm for continual, personalized human-computer interaction optimization.

Technology Category

• Humans and AI: Human-in-the-loop Machine Learning
• Search and Optimization: Learning to Search
• Machine Learning: Optimization

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Optimal input settings vary across users due to differences in motor abilities and personal preferences, which are typically addressed by manual tuning or calibration. Although human-in-the-loop optimization has the potential to identify optimal settings during use, it is rarely applied due to its long optimization process. A more efficient approach would continually leverage data from previous users to accelerate optimization, exploiting shared traits while adapting to individual characteristics. We introduce the concept of Continual Human-in-the-Loop Optimization and a Bayesian optimization-based method that leverages a Bayesian-neural-network surrogate model to capture population-level characteristics while adapting to new users. We propose a generative replay strategy to mitigate catastrophic forgetting. We demonstrate our method by optimizing virtual reality keyboard parameters for text entry using direct touch, showing reduced adaptation times with a growing user base. Our method opens the door for next-generation personalized input systems that improve with accumulated experience.

Problem

• Optimize input settings for diverse user abilities and preferences.
• Reduce long optimization times in human-in-the-loop systems.
• Enable personalized input systems that improve with user experience.

Innovation

• Bayesian-neural-network surrogate model for optimization
• Generative replay strategy to prevent forgetting
• Continual Human-in-the-Loop Optimization for personalization