Mahrokh Ghoddousi Boroujeni

Mahrokh Ghoddousi Boroujeni
Mahrokh Ghoddousi Boroujeni
PhD Student
I want to leave Earth a better place by contributing to science and the environment, and NCCR Automation is the way for me to do both.

Mahrokh Ghoddousi is a doctoral assistant at DECODE, EPFL, jointly supervised by Prof. Giancarlo Ferrari Trecate from EPFL and Prof. Andreas Krause from ETHZ. Mahrokh holds two BSc degrees in Electrical Engineering and Computer Science from Sharif University (Iran). She previously did research internships at IFA (ETHZ), PVLab (EPFL), and Movement Generation Lab (Max Planck Institute).

Mahrokh has taken part in the #NCCRWomen campaign. You can see her great video here

Research projects as Researcher

Titolo
Principal Investigators

Distributed GP learning and model-based RL

Sommario

The goal of the project is to study data-driven modelling and control frameworks for large-scale systems stemming from the interconnection of several subsystems, which are of primary interest to the NCCR. For this purpose, we will develop (i) innovative probabilistic machine learning algorithms, based on Gaussian Processes (GPs), for batch and online estimation of models complying with the coupling structure of the system and (ii) model-based Distributed Reinforcement Learning (DRL) algorithms for the generation of networked control policies. Our DRL schemes will exploit the epistemic model uncertainty to drive exploration while ensuring stability of the distributed controllers.

Distributed GP learning and model-based RL

The goal of the project is to study data-driven modelling and control frameworks for large-scale systems stemming from the interconnection of several subsystems, which are of primary interest to the NCCR. For this purpose, we will develop (i) innovative probabilistic machine learning algorithms, based on Gaussian Processes (GPs), for batch and online estimation of models complying with the coupling structure of the system and (ii) model-based Distributed Reinforcement Learning (DRL) algorithms for the generation of networked control policies. Our DRL schemes will exploit the epistemic model uncertainty to drive exploration while ensuring stability of the distributed controllers.

110
63103095-1b2c-42bc-96bf-4e514b510220