← Back to all news
arXiv Statistics ML By Boning Zhang, Matteo Zecchin, Mingzhao Guo, Dongzhu Liu, Osvaldo Simeone

Federated Martingale Posterior Samping

Read the original on arXiv Statistics ML →

The paper introduces Federated Martingale Posterior (FMP) sampling, a one‑shot, embarrassingly parallel method for federated Bayesian neural networks. FMP replaces the traditional prior–likelihood framework with a predictive distribution, allowing each client to upload only a small set of trainable data embeddings while the server performs predictive sampling centrally. Experiments demonstrate that FMP closely matches the performance of a centralized approach and achieves the lowest mean expected calibration error among evaluated one‑shot federated methods.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Statistics ML.

Related stories

arXiv AI
Oct 9

When Has a Bayesian Neural Network Sampled Enough? Adaptive Inference Time with Statistical Guarantees

The paper introduces a method that uses confidence sequences to adaptively determine the number of Monte Carlo samples needed for Bayesian neural network predictions, ensuring statistical guarantees. It applies to tasks such as class identification, full predictive distribution approximation, and probability-threshold decisions, stopping sampling once the desired confidence is achieved. Experiments demonstrate efficient computational budgeting, allocating more samples to ambiguous inputs, preserving decision reliability, and reducing overall latency compared to a fixed sample budget.

By Fabian Denoodt, Sibylle Hess
More like this →
arXiv Machine Learning
Sep 7

PACE: Propagation-Aware Collaborative Correction for One-Shot Personalized Federated Graph Learning

PACE introduces a propagation‑aware collaborative correction for one‑shot personalized federated graph learning. Each client sends a rank‑r update and a diagonal sketch of message moments, allowing the server to construct a correction that anchors to the receiver’s local model. A convex negative‑log‑likelihood calibration selects a single coefficient to blend local and external logits, improving accuracy and weighted‑F1 on most datasets while preserving local predictions when the correction is unhelpful.

By Ruizhe Huang, Chengran Li, Xiaochuan Shi
More like this →