Grad-FEC: Unequal Loss Protection of Deep Features in Collaborative Intelligence
Collaborative intelligence (CI) involves dividing an artificial intelligence (AI) model into two parts: front-end, to be deployed on an edge device, and back-end, to be deployed in the cloud. The deep feature tensors produced by the front-end are transmitted to the cloud through a communication channel, which may be subject to packet loss. To address this issue, in this paper, we propose a novel approach to enhance the resilience of the CI system in the presence of packet loss through Unequal Loss Protection (ULP). The proposed ULP approach involves a feature importance estimator, which estimates the importance of feature packets produced by the front-end, and then selectively applies Forward Error Correction (FEC) codes to protect important packets. Experimental results demonstrate that the proposed approach can significantly improve the reliability and robustness of the CI system in the presence of packet loss.
Code (0)
등록된 구현이 없습니다.
Tasks
Feature ImportanceSimilar Papers 제목 키워드 기반
A Video-Aware FEC-Based Unequal Loss Protection System for Video Streaming over RTP
A video-aware unequal loss protection (ULP) system for protecting RTP video streaming in bursty packet loss networks is proposed. Considering the relevance of the frame, the state of the channel, and the bitrate constrai…
Multicarrier Modulation-Based Digital Radio-over-Fibre System Achieving Unequal Bit Protection with Over 10 dB SNR Gain
We propose a multicarrier modulation-based digital radio-over-fibre system achieving unequal bit protection by bit and power allocation for subcarriers. A theoretical SNR gain of 16.1 dB is obtained in the AWGN channel a…
Forward Error Correction applied to JPEG-XS codestreams
JPEG-XS offers low complexity image compression for applications with constrained but reasonable bit-rate, and low latency. Our paper explores the deployment of JPEG-XS on lossy packet networks. To preserve low latency, …
Image CompressionPolar Code Based Federated Learning: Convergence Analysis and Resource Allocation
Federated learning (FL) enables collaborative model training across distributed devices without sharing raw data; however, it faces significant communication bottlenecks and channel impairments in practice. Conventional …
Federated LearningFrom Bit-Position Sensitivity to Unequal Error Protection for DNN Inference Memory
We characterize per-bit-position fault sensitivity in ML inference across 16 workloads -- spanning transformer-based models and attention-free CNNs -- and across three floating-point formats. Our central empirical findin…