TY - GEN
T1 - Symbolic Regressor
T2 - 20th International Conference on Hybrid Artificial Intelligence Systems, HAIS 2025
AU - Rey-Arnal, Danel
AU - G. Bringas, Pablo
AU - Laña, Ibai
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2026
Y1 - 2026
N2 - Multiple sources of worries such as economic constraints and the dangers of climate change have moved society towards the process of optimizing the use of their electricity. However this approach towards energy consumption has become a source of uncertainty and worry as load monitoring becomes the norm. In order to overcome the privacy concerns techniques on Non-Intrusive Load Monitoring have been in development since the 1980s. In the field of load disaggregation applications of NILM there is constant reference to three topics to be improved on, results, interpretability and responsiveness. This paper investigates the role symbolic regression tools in the field of NILM, both as a singular tool of disaggregation and as a support instrument of deep learning models more common in the literature, such as LSTM, to improve on their prediction capabilities and adding a layer of interpretability to the results. The experimentation of this document offer two different solutions with various degrees of success depending on the proposed scenario although with quantifiable improvement over the established baseline.
AB - Multiple sources of worries such as economic constraints and the dangers of climate change have moved society towards the process of optimizing the use of their electricity. However this approach towards energy consumption has become a source of uncertainty and worry as load monitoring becomes the norm. In order to overcome the privacy concerns techniques on Non-Intrusive Load Monitoring have been in development since the 1980s. In the field of load disaggregation applications of NILM there is constant reference to three topics to be improved on, results, interpretability and responsiveness. This paper investigates the role symbolic regression tools in the field of NILM, both as a singular tool of disaggregation and as a support instrument of deep learning models more common in the literature, such as LSTM, to improve on their prediction capabilities and adding a layer of interpretability to the results. The experimentation of this document offer two different solutions with various degrees of success depending on the proposed scenario although with quantifiable improvement over the established baseline.
KW - load disaggregation
KW - LSTM
KW - NILM
KW - Symbolic Regression
UR - https://www.scopus.com/pages/publications/105020675781
U2 - 10.1007/978-3-032-08465-1_21
DO - 10.1007/978-3-032-08465-1_21
M3 - Conference contribution
AN - SCOPUS:105020675781
SN - 9783032084644
T3 - Lecture Notes in Computer Science
SP - 260
EP - 271
BT - Hybrid Artificial Intelligent Systems - 20th International Conference, HAIS 2025, Proceedings
A2 - Corchado, Emilio
A2 - Quintián, Héctor
A2 - Troncoso Lora, Alicia
A2 - Pérez García, Hilde
A2 - Jove Pérez, Esteban
A2 - Calvo Rolle, José Luis
A2 - Martínez de Pisón, Francisco Javier
A2 - García Bringas, Pablo
A2 - Martínez Álvarez, Francisco
A2 - Herrero, Álvaro
A2 - Fosci, Paolo
A2 - Sérgio Filipe, Ramos
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 16 October 2025 through 17 October 2025
ER -