Personalized federated learning for buildings energy consumption forecasting
Bibliographic record
Abstract
Buildings' energy consumption forecasting is critical for energy saving and building maintenance. However, most studies only focus on centralized learning of one dataset, which ignores the data privacy and data shortage issue. Meanwhile, the difference in energy data distributions from many buildings causes difficulties in training a good machine learning model. Although these two challenges of data privacy and data heterogeneity could be resolved through personalized federated learning algorithms to some degree, there is still a lack of investigation into applying these algorithms to building energy data analytics. Besides using existing personalized federated learning algorithms, we design a new deep learning model through a mixture of experts to support personalization for heterogeneous data distribution. This new design is the first trial to tackle the data heterogeneity through ensemble architecture in federated load forecasting. Extensive experiments are conducted to evaluate the effectiveness of our proposed model with different training algorithms. The results show that our proposed method outperforms other state-of-the-art models in energy forecasting accuracies by 10% to 40% across the buildings' energy data from university campuses. • Create a new personalized federated learning algorithm through deep learning architecture model design. • Develop a new transformer-based deep learning model called Metaformer, which reduces memory usage by 50%. • Improve buildings' energy forecasting accuracy by 10% to 40% compared to state-of-the-art models across 40 buildings.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".