Research map / Federated learning and privacy / Communication-efficient FL
BIQ: Bisection Interval Quantization for Communication-efficient Federated Learning
Luyang Gai, Shusen Yang, Xuebin Ren, Zihao Zhou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace stochastic quantization (QSGD-based FedPAQ) with bisection interval quantization | FedAvg with 3-bit quantization · CIFAR-10 (I.I.D.) | accuracy %: 70.83 -> 77.49 | Improved |
| replace stochastic quantization with BIQ | FedAvg with 3-bit quantization · Tiny ImageNet (I.I.D.) | top-1 accuracy %: 18.19 -> 20.29 | Improved |
| use WBIQ weighted reconstruction instead of BIQ midpoint | FedBIQ · CIFAR-10 (I.I.D.) | accuracy %: 77.49 -> 78.06 | Improved |
| use WBIQ instead of BIQ | FedBIQ · Tiny ImageNet (I.I.D.) | top-1 accuracy %: 20.29 -> 20.39 | No clear effect |
| use BIQ instead of non-uniform FedNQFL | quantized FL · MNIST | avg cost per round: 12.05 -> 9.47 | Improved |
| use BIQ instead of non-uniform FedNQFL | quantized FL · Tiny ImageNet (I.I.D.) | top-1 accuracy %: 5.92 -> 20.29 | Improved |
| quantize with 3-bit BIQ instead of 32-bit FedAvg | FedAvg · MNIST (I.I.D.) | accuracy %: 95.58 -> 95.22 | No clear effect |
| add quantization with sparsification (FedSQFL) | FedSQ-style quantization · CIFAR-10 (I.I.D.) | accuracy %: 70.67 -> 66.56 | Hurt |
Closest papers by meaning
- Mixed-Precision Quantization for Federated Learning on Resource-Constrained Heterogeneous Devices (CVPR 2024 · similarity 0.88)
- LBI-FL: Low-Bit Integerized Federated Learning with Temporally Dynamic Bit-Width Allocation (ICML 2025 · similarity 0.87)
- Fed-QSSL: A Framework for Personalized Federated Learning under Bitwidth and Data Heterogeneity (AAAI 2024 · similarity 0.85)
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 2025 · similarity 0.85)
- FedWMSAM: Fast and Flat Federated Learning via Weighted Momentum and Sharpness-Aware Minimization (NeurIPS 2025 · similarity 0.84)
- Layer-wise Update Aggregation with Recycling for Communication-Efficient Federated Learning (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
