Get Started
Home
Topics
Search
Library
Research questionHow can zero-shot building IoT forecasting adapt covariates to each building’s physical topology?Building sensors are physically coupled and influenced by operational context, but forecasting systems often treat them as isolated series or use the same covariates everywhere. This makes it difficult to transfer forecasts reliably when buildings have different layouts, systems, and sensor relationships.
AI
Machine Learning
Research Paper
Statistical Machine Learning
Technology
Latest papersRecent research connected to this question, newest first.TopoBrick: Agentic Topology Sampling of Exogenous Variables for Zero-Shot Building IoT ForecastingThe source concerns training-free forecasting for building IoT sensors using building knowledge graphs. It distinguishes past-known sensor states from future-known calendar, schedule, and meteorological variables, and reports results across three real-world buildings, including HVAC- and weather-linked sensing variables.research paper · Sep 2, 2026
Related questions
How can online time-series forecasters adapt to evolving, unseen patterns without costly continual parameter updates?How can we detect multivariate IoT traffic anomalies across deployments without labels or raw-feature sharing?How can lightweight time-series forecasters learn accurate, specialized forecasts from scarce private data?How can zero-shot LLMs predict weather-related distribution-grid outage severity without labeled data?