A platform for research: civil engineering, architecture and urbanism
Action Prediction in Smart Home Based on Reinforcement Learning
This paper presents an “intelligent” environment that can be occupied by an elderly or handicapped person. It is characterized by its online learning and continuous adaptation based on a new algorithm called “Planning Q-learning Algorithm (PQLA)”. The user can make feedback promptly which simulates an algorithm that reconfigures the existing plans. The software adaptation is run under middleware “WCOMP” based on the aspect of assembly concept to adapt to the environmental changes.
Action Prediction in Smart Home Based on Reinforcement Learning
This paper presents an “intelligent” environment that can be occupied by an elderly or handicapped person. It is characterized by its online learning and continuous adaptation based on a new algorithm called “Planning Q-learning Algorithm (PQLA)”. The user can make feedback promptly which simulates an algorithm that reconfigures the existing plans. The software adaptation is run under middleware “WCOMP” based on the aspect of assembly concept to adapt to the environmental changes.
Action Prediction in Smart Home Based on Reinforcement Learning
Lect.Notes Computer
Bodine, Cathy (editor) / Helal, Sumi (editor) / Gu, Tao (editor) / Mokhtari, Mounir (editor) / Hassan, Marwa (author) / Atieh, Mirna (author)
International Conference on Smart Homes and Health Telematics ; 2014 ; Denver, CO, USA
2014-12-27
6 pages
Article/Chapter (Book)
Electronic Resource
English
Ambient computing , Intelligent environment , Online learning , Reinforcement learning , Software adaptation Computer Science , Special Purpose and Application-Based Systems , User Interfaces and Human Computer Interaction , Information Systems Applications (incl. Internet) , Artificial Intelligence , Image Processing and Computer Vision
Action Prediction in Smart Home Based on Reinforcement Learning
British Library Conference Proceedings | 2015
|British Library Conference Proceedings | 2021
|