Packet-Loss-Tolerant Split Inference for Delay-Sensitive Deep Learning in Lossy Wireless Networks

• URL: http://arxiv.org/abs/2104.13629v1
• Date: Wed, 28 Apr 2021 08:28:22 GMT
• Title: Packet-Loss-Tolerant Split Inference for Delay-Sensitive Deep Learning in Lossy Wireless Networks
• Authors: Sohei Itahara, Takayuki Nishio, and Koji Yamamoto
• Abstract summary: In distributed inference, computational tasks are offloaded from the IoT device to other devices or the edge server via lossy IoT networks. narrow-band and lossy IoT networks cause non-negligible packet losses and retransmissions, resulting in non-negligible communication latency. We propose a split inference with no retransmissions (SI-NR) method that achieves high accuracy without any retransmissions, even when packet loss occurs.
• Score: 4.932130498861988
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The distributed inference framework is an emerging technology for real-time applications empowered by cutting-edge deep machine learning (ML) on resource-constrained Internet of things (IoT) devices. In distributed inference, computational tasks are offloaded from the IoT device to other devices or the edge server via lossy IoT networks. However, narrow-band and lossy IoT networks cause non-negligible packet losses and retransmissions, resulting in non-negligible communication latency. This study solves the problem of the incremental retransmission latency caused by packet loss in a lossy IoT network. We propose a split inference with no retransmissions (SI-NR) method that achieves high accuracy without any retransmissions, even when packet loss occurs. In SI-NR, the key idea is to train the ML model by emulating the packet loss by a dropout method, which randomly drops the output of hidden units in a DNN layer. This enables the SI-NR system to obtain robustness against packet losses. Our ML experimental evaluation reveals that SI-NR obtains accurate predictions without packet retransmission at a packet loss rate of 60%.

Related papers

• Towards Resource-Efficient Federated Learning in Industrial IoT for Multivariate Time Series Analysis [50.18156030818883]
  Anomaly and missing data constitute a thorny problem in industrial applications. Deep learning enabled anomaly detection has emerged as a critical direction. The data collected in edge devices contain user privacy.
  arXiv  Detail & Related papers  (2024-11-06T15:38:31Z)
• Adaptive ResNet Architecture for Distributed Inference in Resource-Constrained IoT Systems [7.26437825413781]
  This paper presents an empirical study that identifies the connections in ResNet that can be dropped without significantly impacting the model's performance. Our experiments demonstrate that an adaptive ResNet architecture can reduce shared data, energy consumption, and latency throughout the distribution.
  arXiv  Detail & Related papers  (2023-07-21T11:07:21Z)
• Performance Limits of a Deep Learning-Enabled Text Semantic Communication under Interference [89.91583691993071]
  We study the performance limits of a popular text SemCom system named DeepSC in the presence of (multi-interferer) RFI. We show that DeepSC produces semantically irrelevant sentences as the power of (multi-interferer) RFI gets very large. We also derive DeepSC's practical limits and a lower bound on its outage probability under multi-interferer RFI.
  arXiv  Detail & Related papers  (2023-02-15T05:43:08Z)
• SlimFL: Federated Learning with Superposition Coding over Slimmable Neural Networks [56.68149211499535]
  Federated learning (FL) is a key enabler for efficient communication and computing leveraging devices' distributed computing capabilities. This paper proposes a novel learning framework by integrating FL and width-adjustable slimmable neural networks (SNNs) We propose a communication and energy-efficient SNN-based FL (named SlimFL) that jointly utilizes superposition coding (SC) for global model aggregation and superposition training (ST) for updating local models.
  arXiv  Detail & Related papers  (2022-03-26T15:06:13Z)
• Communication-oriented Model Fine-tuning for Packet-loss Resilient Distributed Inference under Highly Lossy IoT Networks [6.107812768939554]
  distributed inference (DI) is a technique for real-time applications empowered by cutting-edge deep machine learning (ML) on resource-constrained Internet of things (IoT) devices. In DI, computational tasks are offloaded from the IoT device to the edge server via lossy IoT networks. We propose a communication-oriented model tuning (COMtune) to achieve highly accurate DI with low-latency but unreliable communication links.
  arXiv  Detail & Related papers  (2021-12-17T09:40:21Z)
• Computational Intelligence and Deep Learning for Next-Generation Edge-Enabled Industrial IoT [51.68933585002123]
  We investigate how to deploy computational intelligence and deep learning (DL) in edge-enabled industrial IoT networks. In this paper, we propose a novel multi-exit-based federated edge learning (ME-FEEL) framework. In particular, the proposed ME-FEEL can achieve an accuracy gain up to 32.7% in the industrial IoT networks with the severely limited resources.
  arXiv  Detail & Related papers  (2021-10-28T08:14:57Z)
• Unit-Modulus Wireless Federated Learning Via Penalty Alternating Minimization [64.76619508293966]
  Wireless federated learning (FL) is an emerging machine learning paradigm that trains a global parametric model from distributed datasets via wireless communications. This paper proposes a wireless FL framework, which uploads local model parameters and computes global model parameters via wireless communications.
  arXiv  Detail & Related papers  (2021-08-31T08:19:54Z)
• Loss Tolerant Federated Learning [6.595005044268588]
  In this paper, we explore the loss tolerant federated learning (LT-FL) in terms of aggregation, fairness, and personalization. We use ThrowRightAway (TRA) to accelerate the data uploading for low-bandwidth-devices by intentionally ignoring some packet losses. The results suggest that, with proper integration, TRA and other algorithms can together guarantee the personalization and fairness performance in the face of packet loss below a certain fraction.
  arXiv  Detail & Related papers  (2021-05-08T04:44:47Z)
• Optimizing Resource-Efficiency for Federated Edge Intelligence in IoT Networks [96.24723959137218]
  We study an edge intelligence-based IoT network in which a set of edge servers learn a shared model using federated learning (FL) We propose a novel framework, called federated edge intelligence (FEI), that allows edge servers to evaluate the required number of data samples according to the energy cost of the IoT network. We prove that our proposed algorithm does not cause any data leakage nor disclose any topological information of the IoT network.
  arXiv  Detail & Related papers  (2020-11-25T12:51:59Z)
• SoS-RPL: Securing Internet of Things Against Sinkhole Attack Using RPL Protocol-Based Node Rating and Ranking Mechanism [1.2691047660244335]
  IoTs are endowed with particular routing disobedience called sinkhole attack owing to their distributed features. In these attacks, a malicious node broadcasts illusive information regarding the routings to impose itself as a route towards specific nodes for the neighboring nodes and thus, attract data traffic. In this paper, the technique is assessed through wide simulations performed within the NS-3 environment. Based on the results of the simulation, it is indicated that the IoT network behavior metrics are enhanced based on the detection rate, false-negative rate, false-positive rate, packet delivery rate, maximum throughput, and packet loss rate.
  arXiv  Detail & Related papers  (2020-05-17T09:26:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.