Federated Learning with Non-IID Data
Explore this paper's citation graph
Summary
This work presents a strategy to improve training on non-IID data by creating a small subset of data which is globally shared between all the edge devices, and shows that accuracy can be increased by 30% for the CIFAR-10 dataset with only 5% globally shared data.
- Type
- preprint
- Published
- 2018-06-02
- Cited by
- 3,436
- References
- 31
- Access
- Open access
- OpenAlex
- https://openalex.org/W2807006176
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:46936175
Keywords
Computer science, Train, Focus (optics), Enhanced Data Rates for GSM Evolution, Federated learning
References
- The complete works of William Shakespeare : comprising his plays and poems, also the history of his life, his will and an introduction to each play to which is added an index to the characters
- Project Adam: Building an Efficient and Scalable Deep Learning Training System
- Perturbed Iterate Analysis for Asynchronous Stochastic Optimization
- Qualitatively characterizing neural network optimization problems
- Estimation, Optimization, and Parallelism when Data is Sparse
- Gradient-based learning applied to document recognition
- Federated Optimization: Distributed Optimization Beyond the Datacenter
- Fractional Max-Pooling
- Hogwild: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent
- ImageNet classification with deep convolutional neural networks
- Large Scale Distributed Deep Networks
- Revisiting Distributed Synchronous SGD
- Federated Learning: Strategies for Improving Communication Efficiency
- Communication-Efficient Learning of Deep Networks from Decentralized Data
- Practical Secure Aggregation for Privacy-Preserving Machine Learning
- Hello Edge: Keyword Spotting on Microcontrollers
- Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
- CMSIS-NN: Efficient Neural Network Kernels for Arm Cortex-M CPUs
- Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition
- Making Gradient Descent Optimal for Strongly Convex Stochastic Optimization
Cited by
- Client Selection for Federated Learning with Heterogeneous Resources in Mobile Edge
- Multi-Institutional Deep Learning Modeling Without Sharing Patient Data: A Feasibility Study on Brain Tumor Segmentation
- LoAdaBoost: Loss-Based AdaBoost Federated Machine Learning on medical Data
- Partitioned Variational Inference: A unified framework encompassing federated and continual learning
- Communication-Efficient On-Device Machine Learning: Federated Distillation and Augmentation under Non-IID Private Data
- Wireless Network Intelligence at the Edge
- Multi-Objective Evolutionary Federated Learning
- Efficient Training Management for Mobile Crowd-Machine Learning: A Deep Reinforcement Learning Approach
- Federated Learning via Over-the-Air Computation
- Federated Machine Learning
- Federated Collaborative Filtering for Privacy-Preserving Personalized Recommendation System
- Federated Optimization in Heterogeneous Networks
- Gradient Scheduling with Global Momentum for Non-IID Data Distributed Asynchronous Training
- Robust and Communication-Efficient Federated Learning From Non-i.i.d. Data
- Hybrid-FL: Cooperative Learning Mechanism Using Non-IID Data in Wireless Networks
- Decentralized Bayesian Learning over Graphs
- Variational Federated Multi-Task Learning
- Robust Federated Learning in a Heterogeneous Environment
- Pushing the Limits of Gossip-Based Decentralised Machine Learning
- Performance Analysis and Characterization of Training Deep Learning Models on NVIDIA TX2
Related papers
- Key Trains Identifying and its Reasonable Ratios in the High-Speed Rail Timetable
- Communication to and in trains
- Genetic algorithm based on conflict detecting for solving departure time domains of passenger trains
- Capacity for express trains on mixed traffic lines
- Study on Station Power Model using Train Types
- Simulation Model of Belgrade Suburban Passenger Trains Service Using OpenTrack
- Performance of future regional high speed trains
- Study on the Theory and Model of the Optimal Train Formation Plan of Multi-block Trains at Technical Service Stations
- Spread: Decentralized Model Aggregation for Scalable Federated Learning