Reinforcement learning from human feedbackIn machine learning, reinforcement learning from human feedback (RLHF) or reinforcement learning from human preferences is a technique that trains a "reward model" directly from human feedback and uses the model as a reward function to optimize an agent's policy using reinforcement learning (RL) through an optimization algorithm like Proximal Policy Optimization. The reward model is trained in advance to the policy being optimized to predict if a given output is good (high reward) or bad (low reward).
Energy developmentEnergy development is the field of activities focused on obtaining sources of energy from natural resources. These activities include the production of renewable, nuclear, and fossil fuel derived sources of energy, and for the recovery and reuse of energy that would otherwise be wasted. Energy conservation and efficiency measures reduce the demand for energy development, and can have benefits to society with improvements to environmental issues.
Renewable energy commercializationRenewable energy commercialization involves the deployment of three generations of renewable energy technologies dating back more than 100 years. First-generation technologies, which are already mature and economically competitive, include biomass, hydroelectricity, geothermal power and heat. Second-generation technologies are market-ready and are being deployed at the present time; they include solar heating, photovoltaics, wind power, solar thermal power stations, and modern forms of bioenergy.
Multi-agent reinforcement learningMulti-agent reinforcement learning (MARL) is a sub-field of reinforcement learning. It focuses on studying the behavior of multiple learning agents that coexist in a shared environment. Each agent is motivated by its own rewards, and does actions to advance its own interests; in some environments these interests are opposed to the interests of other agents, resulting in complex group dynamics. Multi-agent reinforcement learning is closely related to game theory and especially repeated games, as well as multi-agent systems.
Réponse à la demandeLa réponse à la demande (demand response) comprend, de manière générale, tous les moyens mis en place pour changer le profil de consommation de l'électricité par les utilisateurs et donc la consommation totale électrique. D'après la Federal Energy Regulatory Commission américaine, elle est le changement d'utilisation de l'électricité par le consommateur final par rapport à l'usage habituel en réponse aux changements de prix de l'électricité au cours du temps, ou des remboursements destinés à diminuer l'usage de l'électricité lors de maxima des prix dans le marché de gros ou lors d'instabilité du réseau électrique.
Apprentissage superviséL'apprentissage supervisé (supervised learning en anglais) est une tâche d'apprentissage automatique consistant à apprendre une fonction de prédiction à partir d'exemples annotés, au contraire de l'apprentissage non supervisé. On distingue les problèmes de régression des problèmes de classement. Ainsi, on considère que les problèmes de prédiction d'une variable quantitative sont des problèmes de régression tandis que les problèmes de prédiction d'une variable qualitative sont des problèmes de classification.
Energy systemAn energy system is a system primarily designed to supply energy-services to end-users. The intent behind energy systems is to minimise energy losses to a negligible level, as well as to ensure the efficient use of energy. The IPCC Fifth Assessment Report defines an energy system as "all components related to the production, conversion, delivery, and use of energy". The first two definitions allow for demand-side measures, including daylighting, retrofitted building insulation, and passive solar building design, as well as socio-economic factors, such as aspects of energy demand management and remote work, while the third does not.
100% renewable energy100% renewable energy means getting all energy from renewable resources. The endeavor to use 100% renewable energy for electricity, heating, cooling and transport is motivated by climate change, pollution and other environmental issues, as well as economic and energy security concerns. Shifting the total global primary energy supply to renewable sources requires a transition of the energy system, since most of today's energy is derived from non-renewable fossil fuels.
Feature (machine learning)In machine learning and pattern recognition, a feature is an individual measurable property or characteristic of a phenomenon. Choosing informative, discriminating and independent features is a crucial element of effective algorithms in pattern recognition, classification and regression. Features are usually numeric, but structural features such as strings and graphs are used in syntactic pattern recognition. The concept of "feature" is related to that of explanatory variable used in statistical techniques such as linear regression.
Apprentissage par transfertL'apprentissage par transfert (transfer learning en anglais) est l'un des champs de recherche de l'apprentissage automatique qui vise à transférer des connaissances d'une ou plusieurs tâches sources vers une ou plusieurs tâches cibles. Il peut être vu comme la capacité d’un système à reconnaître et appliquer des connaissances et des compétences, apprises à partir de tâches antérieures, sur de nouvelles tâches ou domaines partageant des similitudes. Adaptation de domaine (domain adaptation en anglais) Catég
