# Federated Learning under Importance Sampling

arXiv: https://arxiv.org/abs/2012.07383 · HTML: https://mlsanity.com/paper/2012.07383/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Dec 2020.

## Built on by (most cited first, 26 of 26)

- [Tackling System and Statistical Heterogeneity for Federated Learning with Adaptive Client Sampling](https://mlsanity.com/paper/2112.11256/) (2021)
- [Federated Learning Under Intermittent Client Availability and Time-Varying Communication Constraints](https://mlsanity.com/paper/2205.06730/) (2022)
- [DELTA: Diverse Client Sampling for Fasting Federated Learning](https://mlsanity.com/paper/2205.13925/) (2022)
- [Non-IID data in Federated Learning: A Survey with Taxonomy, Metrics, Methods, Frameworks and Future Directions](https://mlsanity.com/paper/2411.12377/) (2024)
- [A Graph Federated Architecture with Privacy Preserving Learning](https://mlsanity.com/paper/2104.13215/) (2021)
- [Adaptive Heterogeneous Client Sampling for Federated Learning over Wireless Networks](https://mlsanity.com/paper/2404.13804/) (2024)
- [Client Selection for Generalization in Accelerated Federated Learning: A Multi-Armed Bandit Approach](https://mlsanity.com/paper/2303.10373/) (2023)
- [Incentive Mechanism Design for Unbiased Federated Learning with Randomized Client Participation](https://mlsanity.com/paper/2304.07981/) (2023)
- [FedStale: leveraging stale client updates in federated learning](https://mlsanity.com/paper/2405.04171/) (2024)
- [ISFL: Federated Learning for Non-i.i.d. Data with Local Importance Sampling](https://mlsanity.com/paper/2210.02119/) (2022)
- [Riemannian Federated Learning via Averaging Gradient Streams](https://mlsanity.com/paper/2409.07223/) (2024)
- [Dynamic Scheduling for Federated Edge Learning with Streaming Data](https://mlsanity.com/paper/2305.01238/) (2023)
- [Aggregation Delayed Federated Learning](https://mlsanity.com/paper/2108.07433/) (2021)
- [Collaborative Mean Estimation over Intermittently Connected Networks with Peer-To-Peer Privacy](https://mlsanity.com/paper/2303.00035/) (2023)
- [Privacy Preserving Semi-Decentralized Mean Estimation over Intermittently-Connected Networks](https://mlsanity.com/paper/2406.03766/) (2024)
- [Privatized Graph Federated Learning](https://mlsanity.com/paper/2203.07105/) (2022)
- [On the Fusion Strategies for Federated Decision Making](https://mlsanity.com/paper/2303.06109/) (2023)
- [A Lightweight Method for Tackling Unknown Participation Statistics in Federated Averaging](https://mlsanity.com/paper/2306.03401/) (2023)
- [Adaptive Federated Learning in Heterogeneous Wireless Networks with Independent Sampling](https://mlsanity.com/paper/2402.10097/) (2024)
- [Causal Influence in Federated Edge Inference](https://mlsanity.com/paper/2405.01260/) (2024)
- [PAUSE: Low-Latency and Privacy-Aware Active User Selection for Federated Learning](https://mlsanity.com/paper/2503.13173/) (2025)
- [Adaptive Federated LoRA in Heterogeneous Wireless Networks with Independent Sampling](https://mlsanity.com/paper/2505.23555/) (2025)
- [Uplink Scheduling in Federated Learning: an Importance-Aware Approach via Graph Representation Learning](https://mlsanity.com/paper/2301.11903/) (2023)
- [Digital versus Analog Transmissions for Federated Learning over Wireless Networks](https://mlsanity.com/paper/2402.09657/) (2024)
- [Online Client Scheduling and Resource Allocation for Efficient Federated Edge Learning](https://mlsanity.com/paper/2410.10833/) (2024)
- [Diffusion Learning with Partial Agent Participation and Local Updates](https://mlsanity.com/paper/2505.11307/) (2025)

## Abstract (arXiv)

Federated learning encapsulates distributed learning strategies that are managed by a central unit. Since it relies on using a selected number of agents at each iteration, and since each agent, in turn, taps into its local data, it is only natural to study optimal sampling policies for selecting agents and their data in federated learning implementations. Usually, only uniform sampling schemes are used. However, in this work, we examine the effect of importance sampling and devise schemes for sampling agents and data non-uniformly guided by a performance measure. We find that in schemes involving sampling without replacement, the performance of the resulting architecture is controlled by two factors related to data variability at each agent, and model variability across agents. We illustrate the theoretical findings with experiments on simulated and real data and show the improvement in performance that results from the proposed strategies.