Continual Error Correction on Low-Resource Devices
This addresses the challenge of improving AI reliability on resource-constrained devices for users, though it appears incremental by building on existing error detection and few-shot learning methods.
The paper tackles the problem of AI prediction errors on low-resource devices by enabling users to correct misclassifications through few-shot learning, achieving over 50% error correction in one-shot scenarios on datasets like Food-101 and Flowers-102 with minimal forgetting and computational overhead.
The proliferation of AI models in everyday devices has highlighted a critical challenge: prediction errors that degrade user experience. While existing solutions focus on error detection, they rarely provide efficient correction mechanisms, especially for resource-constrained devices. We present a novel system enabling users to correct AI misclassifications through few-shot learning, requiring minimal computational resources and storage. Our approach combines server-side foundation model training with on-device prototype-based classification, enabling efficient error correction through prototype updates rather than model retraining. The system consists of two key components: (1) a server-side pipeline that leverages knowledge distillation to transfer robust feature representations from foundation models to device-compatible architectures, and (2) a device-side mechanism that enables ultra-efficient error correction through prototype adaptation. We demonstrate our system's effectiveness on both image classification and object detection tasks, achieving over 50% error correction in one-shot scenarios on Food-101 and Flowers-102 datasets while maintaining minimal forgetting (less than 0.02%) and negligible computational overhead. Our implementation, validated through an Android demonstration app, proves the system's practicality in real-world scenarios.