arXiv:2604.07796v2 Announce Type: replace-cross Abstract: In this paper, we study the problem of mean estimation under 1-bit communication constraints. We propose a novel adaptive mean estimator based solely on randomized threshold queries, where each 1-bit outcome indicates whether a given sample exceeds a sequentially chosen threshold. Our estimator is $(\epsilon, \delta)$-PAC for any distribution with a bounded mean $\mu \in [-\lambda, \lambda]$ and a bounded $k$-th central moment $\mathbb{E}[|X-\mu|^k] \le \sigma^k$ for any fixed $k > 1$. Moreover, our sample complexity is order-optimal in
The continuous growth of data and demand for efficient processing across various domains drives the need for optimized estimation techniques under resource constraints.
This research provides a more efficient and robust method for mean estimation with minimal communication, which is crucial for distributed AI systems, IoT, and edge computing.
The proposed order-optimal and $(\epsilon, \delta)$-PAC estimator allows for more accurate and resource-efficient data aggregation and model training in constrained environments.
- · Edge computing providers
- · IoT device manufacturers
- · Machine learning researchers
- · Distributed AI platforms
- · Inefficient communication protocols
- · Systems requiring high-bandwidth data transfer
More accurate and faster training of machine learning models in resource-limited settings.
Reduced energy consumption and increased scalability for large-scale distributed AI deployments.
Accelerated development of autonomous AI agents operating with minimal communication overhead.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG