Sergio Pérez
Sergio Pérez has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Cristina Ibañez-Llano
Repsol Technology Lab , Repsol , Spain
1 shared publication - Fernando Fernández
Universidad Carlos III de Madrid, Leganés, Spain
1 shared publication - Javier Corrochano
Computer Science Department , Universidad Carlos III de Madrid , Spain
1 shared publication - Javier García
Universidad Santiago de Compostela, Santiago de Compostela, Spain
1 shared publication - José Luis Pérez
Computer Science Department , Universidad Carlos III de Madrid , Spain
1 shared publication - José Manuel Moya
LSI -Integrated System Lab ETSIT Universidad Politécnica de Madrid 28040 Madrid Spain
1 shared publication
Latest Publications
- Open Access
- 1 October 2023
- Published by Walter de Gruyter GmbH in Journal of Artificial Intelligence and Soft Computing Research
- Vol. 13 (4) , 273-287
Discrete Uncertainty Quantification For Offline Reinforcement Learning
In many Reinforcement Learning (RL) tasks, the classical online interaction of the learning agent with the environment is impractical, either because such interaction is expensive or dangerous. In these cases, previous gathered data can be used, arising what is typically called Offline RL. However, this type of learning faces a large number of challenges, mostly derived from the fact that exploration/exploitation trade-off is overshadowed. In addition, the historical data is usually biased by the way it was obtained, typically, a sub-optimal controller, producing a distributional shift from historical data and the one required to learn the optimal policy. In this paper, we present a novel approach to deal with the uncertainty risen by the absence or sparse presence of some state-action pairs in the learning data. Our approach is based on shaping the reward perceived from the environment to ensure the task is solved. We present the approach and show that combining it with classic online RL methods make them perform as good as state of the art Offline RL algorithms such as CQL and BCQ. Finally, we show that using our method on top of established offline learning algorithms can improve them.
- 1 October 2023
- Published by Walter de Gruyter GmbH in Journal of Artificial Intelligence and Soft Computing Research
- Vol. 13 (4) , 273-287
Open AccessDiscrete Uncertainty Quantification For Offline Reinforcement Learning
In many Reinforcement Learning (RL) tasks, the classical online interaction of the learning agent with the environment is impractical, either because such interaction is expensive or dangerous. In these cases, previous gathered data can be used, arising what is typically called Offline RL. However, this type of learning faces a large number of challenges, mostly derived from the fact that exploration/exploitation trade-off is overshadowed. In addition, the historical data is usually biased by the way it was obtained, typically, a sub-optimal controller, producing a distributional shift from historical data and the one required to learn the optimal policy. In this paper, we present a novel approach to deal with the uncertainty risen by the absence or sparse presence of some state-action pairs in the learning data. Our approach is based on shaping the reward perceived from the environment to ensure the task is solved. We present the approach and show that combining it with classic online RL methods make them perform as good as state of the art Offline RL algorithms such as CQL and BCQ. Finally, we show that using our method on top of established offline learning algorithms can improve them.
- Open Access
- 1 December 2021
- Published by Elsevier in Future Generation Computer Systems
- Vol. 125, 891-907
Energy-conscious optimization of Edge Computing through Deep Reinforcement Learning and two-phase immersion cooling
- 1 December 2021
- Published by Elsevier in Future Generation Computer Systems
- Vol. 125, 891-907
Open AccessEnergy-conscious optimization of Edge Computing through Deep Reinforcement Learning and two-phase immersion cooling