A Comprehensive Review and a Taxonomy of Edge Machine Learning: Requirements, Paradigms, and Techniques
arXiv:2302.08571 · doi:10.3390/ai4030039
Abstract
The union of Edge Computing (EC) and Artificial Intelligence (AI) has brought forward the Edge AI concept to provide intelligent solutions close to the end-user environment, for privacy preservation, low latency to real-time performance, and resource optimization. Machine Learning (ML), as the most advanced branch of AI in the past few years, has shown encouraging results and applications in the edge environment. Nevertheless, edge-powered ML solutions are more complex to realize due to the joint constraints from both edge computing and AI domains, and the corresponding solutions are expected to be efficient and adapted in technologies such as data processing, model compression, distributed inference, and advanced learning paradigms for Edge ML requirements. Despite the fact that a great deal of the attention garnered by Edge ML is gained in both the academic and industrial communities, we noticed the lack of a complete survey on existing Edge ML technologies to provide a common understanding of this concept. To tackle this, this paper aims at providing a comprehensive taxonomy and a systematic review of Edge ML techniques, focusing on the soft computing aspects of existing paradigms and techniques. We start by identifying the Edge ML requirements driven by the joint constraints. We then extensively survey more than twenty paradigms and techniques along with their representative work, covering two main parts: edge inference, and edge learning. In particular, we analyze how each technique fits into Edge ML by meeting a subset of the identified requirements. We also summarize Edge ML frameworks and open issues to shed light on future directions for Edge ML.
60 pages, 8 figures, 8 tables, AI ISSN: 2673-2688
References in corpus (47)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Overcoming catastrophic forgetting in neural networks
- A Simple Framework for Contrastive Learning of Visual Representations
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Communication-Efficient Learning of Deep Networks from Decentralized Data
- Language Models are Few-Shot Learners
- Deep Visual Domain Adaptation: A Survey
- Convergence of Edge Computing and Deep Learning: A Comprehensive Survey
- Federated Learning for Internet of Things: A Comprehensive Survey
- Domain Generalization: A Survey
- Self-supervised Learning: Generative or Contrastive
- ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
- A Survey on Mobile Edge Networks: Convergence of Computing, Caching and Communications
- QSGD: Communication-Efficient SGD via Gradient Quantization and Encoding
- Generalizing from a Few Examples: A Survey on Few-Shot Learning
- Predicting Parameters in Deep Learning
- Performance evaluation and hyperparameter tuning of statistical and machine-learning models using spatial data
- FedKD: Communication Efficient Federated Learning via Knowledge Distillation
- Efficient Neural Architecture Search via Parameter Sharing
- Self-Supervised Representation Learning: Introduction, Advances and Challenges
- Federated Learning over Wireless Networks: Convergence Analysis and Resource Allocation
- Machine Learning at the Network Edge: A Survey
- Large Scale Adversarial Representation Learning
- A Survey of Deep Meta-Learning
- FedPAQ: A Communication-Efficient Federated Learning Method with Periodic Averaging and Quantization
- Dynamic Channel Pruning: Feature Boosting and Suppression
- Unsupervised Label Noise Modeling and Loss Correction
- Qualitatively characterizing neural network optimization problems
- Meta-Reinforcement Learning of Structured Exploration Strategies
- DSD: Dense-Sparse-Dense Training for Deep Neural Networks
- A comprehensive review of Binary Neural Network
- Model-Based Meta-Reinforcement Learning for Flight with Suspended Payloads
- Explicit Inductive Bias for Transfer Learning with Convolutional Networks
- On-edge Multi-task Transfer Learning: Model and Practice with Data-driven Task Allocation
- Cluster-to-Conquer: A Framework for End-to-End Multi-Instance Learning for Whole Slide Image Classification
- Asynchronous Federated Learning on Heterogeneous Devices: A Survey
- BERT Loses Patience: Fast and Robust Inference with Early Exit
- Divergence-aware Federated Self-Supervised Learning
- What Language Model Architecture and Pretraining Objective Work Best for Zero-Shot Generalization?
- Connecting Low-Loss Subspace for Personalized Federated Learning
- Fixed-point Quantization of Convolutional Neural Networks for Quantized Inference on Embedded Platforms
- Continual Horizontal Federated Learning for Heterogeneous Data
- From Weakly Supervised Learning to Biquality Learning: an Introduction
- BiFSMN: Binary Neural Network for Keyword Spotting
- Even your Teacher Needs Guidance: Ground-Truth Targets Dampen Regularization Imposed by Self-Distillation