# Adapting to Change: A Comparison of Continual and Transfer Learning for Modeling Building Thermal Dynamics under Concept Drifts

arXiv: https://arxiv.org/abs/2508.21615 · HTML: https://mlsanity.com/paper/2508.21615/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 63 references that are in the map.
- First posted on arXiv 29 Aug 2025.

## Builds on (most cited first, 13 of 13)

- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Gradient Episodic Memory for Continual Learning](https://mlsanity.com/paper/1706.08840/) (2017)
- [A Comprehensive Survey of Continual Learning: Theory, Method and Application](https://mlsanity.com/paper/2302.00487/) (2023)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [Continual Learning: Applications and the Road Forward](https://mlsanity.com/paper/2311.11908/) (2023)
- [A study on the plasticity of neural networks](https://mlsanity.com/paper/2106.00042/) (2021)
- [Deep networks for system identification: a Survey](https://mlsanity.com/paper/2301.12832/) (2023)
- [BuildingsBench: A Large-Scale Dataset of 900K Buildings and Benchmark for Short-Term Load Forecasting](https://mlsanity.com/paper/2307.00142/) (2023)
- [CityLearn v2: Energy-flexible, resilient, occupant-centric, and carbon-aware management of grid-interactive communities](https://mlsanity.com/paper/2405.03848/) (2024)
- [GenTL: A General Transfer Learning Model for Building Thermal Dynamics](https://mlsanity.com/paper/2501.13703/) (2025)
- [Generalized Online Transfer Learning for Climate Control in Residential Buildings](https://mlsanity.com/paper/1610.04042/) (2016)
- [BUILDA: A Thermal Building Data Generation Framework for Transfer Learning](https://mlsanity.com/paper/2508.12703/) (2025)
- [Impact of data for forecasting on performance of model predictive control in buildings with smart energy storage](https://mlsanity.com/paper/2402.12539/) (2024)

## Built on by (most cited first, 7 of 7)

- [SLE-FNO: Single-Layer Extensions for Task-Agnostic Continual Learning in Fourier Neural Operators](https://mlsanity.com/paper/2603.20410/) (2026)
- [Thermal-GEMs: Generalized Models for Building Thermal Dynamics](https://mlsanity.com/paper/2604.16443/) (2026)
- [Counter-Dyna: Data-Efficient RL-Based HVAC Control using Counterfactual Building Models](https://mlsanity.com/paper/2605.04555/) (2026)
- [Neural Parameter Estimation of RC Thermal Building Models for Model Predictive Control](https://mlsanity.com/paper/2604.05904/) (2026)
- [BuilDyn: Excitation-Driven Data Generation for Building Thermal Dynamics Modeling and Control](https://mlsanity.com/paper/2605.29849/) (2026)
- [Very Exciting: Zero-Shot Model Predictive Control of Buildings via Excitation-Based Generalized Transfer Learning Models](https://mlsanity.com/paper/2609.12853/) (2026)
- [The Smart Buildings Challenge: Learning Foundation Models for Real-World Optimization at Scale](https://mlsanity.com/paper/2610.18511/) (2026)

## Abstract (arXiv)

Transfer Learning (TL) is currently the most effective approach for modeling building thermal dynamics when only limited data are available. TL uses a pretrained model that is fine-tuned to a specific target building. However, it remains unclear how to proceed after initial fine-tuning, as more operational measurement data are collected over time. This challenge becomes even more complex when the dynamics of the building change, for example, after a retrofit or a change in occupancy. In Machine Learning literature, Continual Learning (CL) methods are used to update models of changing systems. TL approaches can also address this challenge by reusing the pretrained model at each update step and fine-tuning it with new measurement data. A comprehensive study on how to incorporate new measurement data over time to improve prediction accuracy and address the challenges of concept drifts (changes in dynamics) for building thermal dynamics is still missing. Therefore, this study compares several CL and TL strategies, as well as a model trained from scratch, for thermal dynamics modeling during building operation. The methods are evaluated using 5--7 years of simulated data representative of single-family houses in Central Europe, including scenarios with concept drifts from retrofits and changes in occupancy. We propose a CL strategy (Seasonal Memory Learning) that provides greater accuracy improvements than existing CL and TL methods, while maintaining low computational effort. SML outperformed the benchmark of initial fine-tuning by 28.1% without concept drifts and 34.9% with concept drifts.