Communication-Efficient Edge AI: Algorithms and Systems

Abstract

Artificial intelligence (AI) has achieved remarkable breakthroughs in a wide range of fields, ranging from speech processing, image classification to drug discovery. This is driven by the explosive growth of data, advances in machine learning (especially deep learning), and easy access to vastly powerful computing resources. Particularly, the wide scale deployment of edge devices (e.g., IoT devices) generates an unprecedented scale of data, which provides the opportunity to derive accurate models and develop various intelligent applications at the network edge. However, such enormous data cannot all be sent from end devices to the cloud for processing, due to the varying channel quality, traffic congestion and/or privacy concerns. By pushing inference and training processes of AI models to edge nodes, edge AI has emerged as a promising alternative. AI at the edge requires close cooperation among edge devices, such as smart phones and smart vehicles, and edge servers at the wireless access points and base stations, which however result in heavy communication overheads. In this paper, we present a comprehensive survey of the recent developments in various techniques for overcoming these communication challenges. Specifically, we first identify key communication challenges in edge AI systems. We then introduce communication-efficient techniques, from both algorithmic and system perspectives for training and inference tasks at the network edge. Potential future research directions are also highlighted.

arXiv commentThis work has been submitted to the IEEE for possible publication

References in corpus (56)

• Communication-Efficient Learning of Deep Networks from Decentralized Data
• A Brief Survey of Deep Reinforcement Learning
• Federated Learning: Strategies for Improving Communication Efficiency
• Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
• Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
• BinaryConnect: Training Deep Neural Networks with binary weights during propagations
• ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Models
• Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
• Edge Intelligence: The Confluence of Edge Computing and Artificial Intelligence
• Deep Learning with Limited Numerical Precision
• Compressing Deep Convolutional Networks using Vector Quantization
• Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
• Speeding Up Distributed Machine Learning Using Codes
• Federated Optimization:Distributed Optimization Beyond the Datacenter
• Compressing Neural Networks with the Hashing Trick
• Speeding up Convolutional Neural Networks with Low Rank Expansions
• Horovod: fast and easy distributed deep learning in TensorFlow
• Federated Optimization in Heterogeneous Networks
• Private federated learning on vertically partitioned data via entity resolution and additively homomorphic encryption
• Communication Efficient Distributed Optimization using an Approximate Newton-type Method
• Bayesian Compression for Deep Learning
• Device Placement Optimization with Reinforcement Learning
• Communication Compression for Decentralized Training
• SecureBoost: A Lossless Federated Learning Framework
• A Linearly-Convergent Stochastic L-BFGS Algorithm
• Cooperative SGD: A unified Framework for the Design and Analysis of Communication-Efficient SGD Algorithms
• Distributed Mean Estimation with Limited Communication
• High-Dimensional Stochastic Gradient Quantization for Communication-Efficient Edge Learning
• Federated Multi-Task Learning
• signSGD: Compressed Optimisation for Non-Convex Problems
• GossipGraD: Scalable Deep Learning using Gossip Communication based Asynchronous Gradient Descent
• AdaComp : Adaptive Residual Gradient Compression for Data-Parallel Distributed Training
• A Quasi-Newton Method Based Vertical Federated Learning Framework for Logistic Regression
• Reconfigurable-Intelligent-Surface Empowered Wireless Communications: Challenges and Opportunities
• Communication Efficient Federated Learning over Multiple Access Channels
• Optimizing Network Performance for Distributed DNN Training on GPU Clusters: ImageNet/AlexNet Training in 1.5 Minutes
• An exploration of parameter redundancy in deep networks with circulant projections
• Data Shuffling in Wireless Distributed Computing via Low-Rank Optimization
• Scalable Deep Learning on Distributed Infrastructures: Challenges, Techniques and Tools
• Learning a Code: Machine Learning for Approximate Non-Linear Coded Computation
• Gossip training for deep learning
• Adaptive Communication Strategies to Achieve the Best Error-Runtime Trade-off in Local-Update SGD
• Iterative Hessian sketch: Fast and accurate solution approximation for constrained least-squares
• GIANT: Globally Improved Approximate Newton Method for Distributed Optimization
• Not Just Privacy: Improving Performance of Private Deep Learning in Mobile Cloud
• Client-Edge-Cloud Hierarchical Federated Learning
• Reconfigurable Intelligent Surface for Green Edge Inference
• Communication-Efficient Distributed Strongly Convex Stochastic Optimization: Non-Asymptotic Rates
• Distributed Sensing with Orthogonal Multiple Access: To code or not to Code?
• Energy-Efficient Processing and Robust Wireless Cooperative Transmission for Edge Inference
• Distributed Stochastic Gradient Descent Using LDGM Codes
• Towards a Theoretical Understanding of Hashing-Based Neural Nets
• Communication constrained cloud-based long-term visual localization in real time
• Elastic Gossip: Distributing Neural Network Training Using Gossip-like Protocols
• Distributed Deep Learning Strategies For Automatic Speech Recognition
• Learning Compact Neural Networks with Regularization

Cited by in corpus (5)

• Wireless for Machine Learning
• Communication-Computation Trade-Off in Resource-Constrained Edge Inference
• Machine Learning for Predictive Deployment of UAVs with Multiple Access
• Branchy-GNN: a Device-Edge Co-Inference Framework for Efficient Point Cloud Processing
• Robust error bounds for quantised and pruned neural networks