We have a new paper out on using generative adversarial networks to study the robustness of RL agents against uncertainties. We show how we can leverage past data to create a new dataset that helps us judge this. https://www.sciencedirect.com/science/article/pii/S0378778821006186?via%3Dihub F. Read more
IEL @ ACM BuildSys’20
We were well represented at ACM BuildSys this year (http://buildsys.acm.org/2020/). See the presentations on YouTube: Jose kicked us off with a full paper presentation MARLISA: Multi-Agent Reinforcement Learning with Iterative Sequential Action Selection for Load Shaping of Grid-Interactive Read more
The good, the bad, and the ugly: Data-driven load profile discord identification in a large building portfolio
Our paper has been accepted in Energy and Buildings. The work is led by our awesome PhD student June, in collaboration with the National Renewable Energy Laboratory in Golden, CO. Check it out y'all. https://doi.org/10.1016/j.enbuild.2020.109892 Abstract: Reducing the overall energy Read more
Wireless sensor network for estimating building performance
Our latest paper, led by colleagues from ETH Zurich, deals with developing and deploying wireless sensor networks for building energy monitoring. https://doi.org/10.1016/j.autcon.2019.103043 Abstract Accurate building energy assessments are often limited by inaccurate assumptions Read more
A critical review of field implementations of occupant-centric building controls
Our review paper on practical implementations of OCC applications is published in Building & Environment. This study is a first output of Subtask 4 of the IEA-EBC Annex 79 research project that is co-lead by IEL. Check it out this international collaboration between groups in Canada, Denmark, Read more
- 1
- 2
- 3
- …
- 6
- Next Page »
