Research Projects

Research Projects, Consortia and Scientific Initiatives in which I have collaborated.

CRAI Logo
Timeline20232026My Role
Junior Researcher

CRAI

National

Centre For Responsible AI

CoordinatorBernardete Ribeiro
FundingPRR - Plano de Recuperação e Resiliência - Agenda Mobilizadora para a Inovação Empresarial
Total Budget€ 78M
Synopsis

The aim of CRAI is to find solutions in the area of "responsible artificial intelligence" by establishing a strategy based on three fundamental pillars - explainability, transparency and sustainability, thus aiming to respond to the risks of current machine learning technologies, such as , the bias and lack of transparency of the generated models. The participation of CISUC will include the development of algorithms that allow the creation of transparent, fairer and more reliable models, incorporating techniques of explainability, elimination of bias and robustness, necessary to the market establishment of responsible AI products. CISUC will also develop reliability mechanisms in human-computer interaction, as well as study the energy efficiency of the models, leading to a more sustainable AI. The consortium will mobilize companies, hospitals and research centers across the country, comprising 11 startups (Unbabel, Feedzai, Sword Health, Apres, Automaise, Emotai, NeuralShift, Priberam, Visor.ai, YData and YooniK), 8 centers research centers located in Lisbon, Porto and Coimbra (Champalimaud Foundation, CISUC, Fraunhoffer Portugal AICOS, FEUP, INESC-ID, IST and IT) and several partners in the health area (Bial, Centro Hospitalar São João, Luz Saúde), from area of Tourism (Pestana Group) and Retail (Sonae).

My Contributions

Developement of sustainable and energy-efficient AI models, focusing on the optimization of machine learning algorithms to reduce computational costs and energy consumption with Spiking Neural Networks and Neuromorphic Computing.

Adaptative Computation
View Project
Smart5Grid Logo
Timeline20212023My Role
Junior Researcher

Smart5Grid

National

Automated 5G Networks and Services for Smart Grids

CoordinatorPaulo Simões
FundingP2020
Total Budget€ 1.36M
Synopsis

Currently the key players of the energy sector, namely the Transmission System Operators (TSOs) and the Distribution System Operators (DSOs) face a great amount of challenges in regards to the delivered energy quality. The ever higher demand of consumers and regulators, the energy liberalisation, the adjustment for the access to new technologies and climate change pressure to provide clean energy are some of the factors that require actions. Consequently, it becomes necessary to provide a higher amount of resiliency, flexibility and security of the energy networks, ultimately increasing operation efficiency. Smart5Grid aims to profit from the advantages of using 5G networks by energy operators (TSOs and DSOs). As such, is is based on three main pillars. The first objective consists on having a 5G network management platform with the utilization of Non-Public Networks (NPNs) in order to provide high levels of reliability and availability and to support thousands of devices per km2 without degradation. In order to provide security to critical systems, such as Smart Grids, the project will also include a security system developed to ensure the protection of control processes inherent to the grid operation, as well as M2M communications, with the possibility of supporting novel protection mechanisms aiming at taking advantage of 5G capabilities. Finally, in the perspective of closing the business loop, a monitoring platform will be provided enabling, among other functionalities, the visualization and analysis of the energy quality.

My Contributions

Developement an a Intrusion Detection System (IDS) for Smart Grid substations with Federated Learning and data analysis techniques to detect cyberattacks in the communication network of the substation

Networks, Communications and Security
Cognitive and Media Systems
View Project