Conformal Distributed Remote Inference in Sensor Networks Under Reliability and Communication Constraints

2024-09-12

arXiv.org

AI Summary

This work addresses inference in multi-label remote classification (e.g., image segmentation) over communication-constrained distributed sensor networks, where strict control of the false negative rate (FNR) under low bandwidth is required. We propose the first dynamic two-tier thresholding framework that integrates online conformal risk control with exponential gradient descent to enable distributed collaborative decision-making: the upper tier employs conformal prediction to guarantee FNR ≤ α in worst-case settings, while the lower tier adaptively optimizes the false positive rate (FPR) via exponential gradient descent and derives its regret bound. Theoretically, our method simultaneously satisfies the FNR constraint and an upper bound on communication cost. Simulation results demonstrate that, under significantly reduced communication overhead, the FNR strictly meets the target α, and the FPR regret remains bounded—validating the framework’s robustness and practicality for edge intelligence applications.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Distributed Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

This paper presents communication-constrained distributed conformal risk control (CD-CRC) framework, a novel decision-making framework for sensor networks under communication constraints. Targeting multi-label classification problems, such as segmentation, CD-CRC dynamically adjusts local and global thresholds used to identify significant labels with the goal of ensuring a target false negative rate (FNR), while adhering to communication capacity limits. CD-CRC builds on online exponentiated gradient descent to estimate the relative quality of the observations of different sensors, and on online conformal risk control (CRC) as a mechanism to control local and global thresholds. CD-CRC is proved to offer deterministic worst-case performance guarantees in terms of FNR and communication overhead, while the regret performance in terms of false positive rate (FPR) is characterized as a function of the key hyperparameters. Simulation results highlight the effectiveness of CD-CRC, particularly in communication resource-constrained environments, making it a valuable tool for enhancing the performance and reliability of distributed sensor networks.

Problem

• Distributed Data Analysis
• Sensor Networks
• Classification Tasks

Innovation

• CD-CRC framework
• Adaptive adjustment
• Online exponential gradient descent