让边缘模型学会辅助老旧设备，提升推理效率。
Learning To Help: Training Models to Assist Legacy Devices
- 设计反向训练框架，让边缘模型学会辅助固定设备
- 找到贝叶斯最优拒绝规则，实测优于传统置信度方法
- 适合资源受限设备的长期部署场景
在物理设备上部署的机器学习模型可能因计算能力有限而过时。由于模型规模大，将部分计算（如推理）卸载到边缘云可帮助这些遗留设备。本文将此问题置于学习中弃权（LWA）框架下，其中边缘端需被训练以协助客户端（设备）。以往工作假设边缘是已知的专家或人类，而本研究则提出反向问题：为固定（遗留）客户端训练边缘专家。如同LWA，客户端使用拒绝规则决定何时将推理任务卸载至边缘（产生代价）。本文推导出贝叶斯最优规则，证明了泛化界，并找到一个一致的代理损失函数。实验结果表明，该框架性能优于基于置信度的拒绝规则。
原文摘要 · Abstract (English)
Machine learning models implemented in hardware on physical devices may be deployed for a long time. The computational abilities of the device may be limited and become outdated with respect to newer improvements. Because of the size of ML models, offloading some computation (e.g. to an edge cloud) can help such legacy devices. We cast this problem in the framework of learning with abstention (LWA) in which the expert (edge) must be trained to assist the client (device). Prior work on LWA trains the client assuming the edge is either an oracle or a human expert. In this work, we formalize the reverse problem of training the expert for a fixed (legacy) client. As in LWA, the client uses a rejection rule to decide when to offload inference to the expert (at a cost). We find the Bayes-optimal rule, prove a generalization bound, and find a consistent surrogate loss function. Empirical results show that our framework outperforms confidence-based rejection rules.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。