OPT-NILM: An Iterative Prior-to-Full-Training Pruning Approach for Cost-Effective User Side Energy Disaggregation.
Sotirios AthanasouliasStavros SykiotisMaria KaselimiAnastasios DoulamisNikolaos DoulamisNikolaos IpiotisPublished in: IEEE Trans. Consumer Electron. (2024)
Keyphrases
- cost effective
- low cost
- cost effectiveness
- energy saving
- training set
- energy consumption
- user interaction
- user interface
- end users
- supervised learning
- search space
- search engine
- avoid overfitting
- user satisfaction
- user model
- user preferences
- traveling salesman problem
- training samples
- energy efficiency
- user experience
- user behavior
- training examples
- real time
- long term
- active learning
- recommender systems
- neural network