The paper introduces an unsupervised peer-relative representation learning framework to detect energy inefficiencies in mobile network sites. It uses a novel energy-aware Minimum Distortion Embedding that incorporates an energy-based repulsion mechanism, creating a low-dimensional space where sites with unusually high consumption are displaced from their peers. The resulting anomaly scores help operators prioritize field investigations, outperforming traditional anomaly detection baselines and supporting large-scale energy-efficiency optimization.
arXiv:2608. 03087v1 Announce Type: cross Abstract: Fine-grained energy consumption data are essential for applications such as demand forecasting, demand response planning, and grid reliability assessment.
By Lin Jiang, Dahai Yu, Ravikumar Gelli, Guang Wang
Fine-grained energy consumption data are essential for applications such as demand forecasting, demand response planning, and grid reliability assessment. However, access to such data is often restricted by privacy concerns and data-sharing constraints, motivating growing interest in synthetic energy data generation.
arXiv:2608. 01426v1 Announce Type: new Abstract: Federated learning (FL) enables distributed optimization and learning across decentralized edge devices while preserving data privacy, but its performance is fundamentally constrained by heterogeneous data distributions, limited communication resources, and energy availability.
By Furkan Bagci, Busra Tegin, Mohammad Kazemi, Tolga M. Duman
arXiv:2608. 09596v1 Announce Type: cross Abstract: Graph Neural Networks (GNNs) suffer from two fundamental limitations: over-smoothing, where node representations become indistinguishable with depth, and over-squashing, where long-range information is compressed through limited message-passing channels.
By Killian Cressant, Pedro B. Velloso
The paper presents a machine learning method to estimate the CPU-package power of individual virtual servers using only metrics observable within the server. Training uses host-side RAPL measurements, while inference relies solely on in‑guest data. Experiments across diverse CPUs, workloads, and configurations show strong predictions when training data match deployment conditions, with some limitations in zero‑shot scenarios, and the approach adds minimal overhead.
By Amandip Sangha