TR2021-149

Meta-Learned Bayesian Optimization for Building Model Calibration using Attentive Neural Processes


    •  Zhan, S., Wichern, G., Laughman, C.R., Chakrabarty, A., "Meta-Learned Bayesian Optimization for Building Model Calibration using Attentive Neural Processes", Advances in Neural Information Processing Systems (NeurIPS), December 2021.
      BibTeX TR2021-149 PDF
      • @inproceedings{Zhan2021dec,
      • author = {Zhan, Sicheng and Wichern, Gordon and Laughman, Christopher R. and Chakrabarty, Ankush},
      • title = {Meta-Learned Bayesian Optimization for Building Model Calibration using Attentive Neural Processes},
      • booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
      • year = 2021,
      • month = dec,
      • url = {https://www.merl.com/publications/TR2021-149}
      • }
  • MERL Contacts:
  • Research Areas:

    Machine Learning, Multi-Physical Modeling, Optimization

Abstract:

Currently, building model calibration algorithms do not leverage data archived from previous related calibration tasks. In this paper, we propose the use of Attentive Neural Processes (ANPs) to meta-learn a distribution across previously seen calibration tasks, which is used to accelerate Bayesian Optimization-based calibration of the unseen target task. Our proposed MetaBOAN algorithm is demonstrated on a library of residential buildings generated by the United States Department of Energy. The experiment results show the significantly improved data efficiency in model calibration.