Helios: Heterogeneity-Aware Federated Learning with Dynamically Balanced Collaboration
Not on the map. Why?
Abstract (arXiv)
In this paper, we propose Helios, a heterogeneity-aware FL framework to tackle the straggler issue. Helios identifies individual devices' heterogeneous training capability, and therefore the expected neural network model training volumes regarding the collaborative training pace. For straggling devices, a "soft-training" method is proposed to dynamically compress the original identical training model into the expected volume through a rotating neuron training approach. With extensive algorithm analysis and optimization schemes, the stragglers can be accelerated while retaining the convergence for local training as well as federated collaboration.
Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). First posted on arXiv 3 Dec 2019.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 5; 2022: 4; 2023: 5; 2024: 1.
15 papers.
Sort:
- 2023Federated Learning for Computationally-Constrained Heterogeneous Devices: A Surveycited by 40
- 2023Every Parameter Matters: Ensuring the Convergence of Federated Learning with Dynamic Heterogeneous Models Reductioncited by 24
- 2021Architectural Patterns for the Design of Federated Learning Systemscited by 17
- 2024Asteroid: Resource-Efficient Hybrid Pipeline Parallelism for Collaborative DNN Training on Heterogeneous Edge Devicescited by 13
- 2021DISTREAL: Distributed Resource-Aware Learning in Heterogeneous Systemscited by 10
- 2021Critical Learning Periods in Federated Learningcited by 8
- 2022Stochastic Coded Federated Learning: Theoretical Analysis and Incentive Mechanism Designcited by 6
- 2023Efficient Federated Learning with Enhanced Privacy via Lottery Ticket Pruning in Edge Computingcited by 5
- 2021Federated Action Recognition on Heterogeneous Embedded Devicescited by 4
- 2022On the Convergence of Heterogeneous Federated Learning with Arbitrary Adaptive Online Model Pruningcited by 4
- 2022FedSS: Federated Learning with Smart Selection of clientscited by 4
- 2022Latency Aware Semi-synchronous Client Selection and Model Aggregation for Wireless Federated Learningcited by 3
- 2024Asteroid: Resource-Efficient Hybrid Pipeline Parallelism for Collaborative DNN Training on Heterogeneous Edge Devices15 Aug 2024
- 2023Every Parameter Matters: Ensuring the Convergence of Federated Learning with Dynamic Heterogeneous Models Reduction12 Oct 2023
- 2023Toward efficient resource utilization at edge nodes in federated learning19 Sep 2023
- 2023Revolutionizing Wireless Networks with Federated Learning: A Comprehensive Review1 Aug 2023
- 2023Federated Learning for Computationally-Constrained Heterogeneous Devices: A Survey18 Jul 2023
- 2023Efficient Federated Learning with Enhanced Privacy via Lottery Ticket Pruning in Edge Computing2 May 2023
- 2022Stochastic Coded Federated Learning: Theoretical Analysis and Incentive Mechanism Design8 Nov 2022
- 2022Latency Aware Semi-synchronous Client Selection and Model Aggregation for Wireless Federated Learning19 Oct 2022
- 2022FedSS: Federated Learning with Smart Selection of clients10 Jul 2022
- 2022On the Convergence of Heterogeneous Federated Learning with Arbitrary Adaptive Online Model Pruning27 Jan 2022
- 2021DISTREAL: Distributed Resource-Aware Learning in Heterogeneous Systems16 Dec 2021
- 2021Nothing Wasted: Full Contribution Enforcement in Federated Edge Learning15 Oct 2021
3 more
- 2021Critical Learning Periods in Federated Learning12 Sep 2021
- 2021Federated Action Recognition on Heterogeneous Embedded Devices18 Jul 2021
- 2021Architectural Patterns for the Design of Federated Learning Systems7 Jan 2021
- 2023Federated Learning for Computationally-Constrained Heterogeneous Devices: A Surveycited by 40
- 2023Every Parameter Matters: Ensuring the Convergence of Federated Learning with Dynamic Heterogeneous Models Reductioncited by 24
- 2021Architectural Patterns for the Design of Federated Learning Systemscited by 17
- 2024Asteroid: Resource-Efficient Hybrid Pipeline Parallelism for Collaborative DNN Training on Heterogeneous Edge Devicescited by 13
- 2021DISTREAL: Distributed Resource-Aware Learning in Heterogeneous Systemscited by 10
- 2021Critical Learning Periods in Federated Learningcited by 8
- 2022Stochastic Coded Federated Learning: Theoretical Analysis and Incentive Mechanism Designcited by 6
- 2023Efficient Federated Learning with Enhanced Privacy via Lottery Ticket Pruning in Edge Computingcited by 5
- 2021Federated Action Recognition on Heterogeneous Embedded Devicescited by 4
- 2022On the Convergence of Heterogeneous Federated Learning with Arbitrary Adaptive Online Model Pruningcited by 4
- 2022FedSS: Federated Learning with Smart Selection of clientscited by 4
- 2022Latency Aware Semi-synchronous Client Selection and Model Aggregation for Wireless Federated Learningcited by 3
12 most cited of 15 All 15 (JSON)
- 2024Asteroid: Resource-Efficient Hybrid Pipeline Parallelism for Collaborative DNN Training on Heterogeneous Edge Devices15 Aug 2024
- 2023Every Parameter Matters: Ensuring the Convergence of Federated Learning with Dynamic Heterogeneous Models Reduction12 Oct 2023
- 2023Toward efficient resource utilization at edge nodes in federated learning19 Sep 2023
- 2023Revolutionizing Wireless Networks with Federated Learning: A Comprehensive Review1 Aug 2023
- 2023Federated Learning for Computationally-Constrained Heterogeneous Devices: A Survey18 Jul 2023
- 2023Efficient Federated Learning with Enhanced Privacy via Lottery Ticket Pruning in Edge Computing2 May 2023
- 2022Stochastic Coded Federated Learning: Theoretical Analysis and Incentive Mechanism Design8 Nov 2022
- 2022Latency Aware Semi-synchronous Client Selection and Model Aggregation for Wireless Federated Learning19 Oct 2022
- 2022FedSS: Federated Learning with Smart Selection of clients10 Jul 2022
- 2022On the Convergence of Heterogeneous Federated Learning with Arbitrary Adaptive Online Model Pruning27 Jan 2022
- 2021DISTREAL: Distributed Resource-Aware Learning in Heterogeneous Systems16 Dec 2021
- 2021Nothing Wasted: Full Contribution Enforcement in Federated Edge Learning15 Oct 2021
12 newest of 15 All 15 (JSON)