PhD & Postdoc Symposium

The Unit organises an annual PhD and Postdoc Symposium for early career researchers to promote their research and get feedback from their peers. 

2023 Symposium

The first PhD and postdoc symposium of the Lisbon ELLIS Unit took place on September 22nd, and was attended by 50 students and postdocs in our network from various fields, including Robotics, Reinforcement Learning, Natural Language Processing, Health, and Computer Vision. 

It featured a talk about the Centre for Responsible AI, given by its CEO Paulo Dimas, a talk given by former PhD student João Santinha on AI and medical imaging, and a panel discussion about "Life after a PhD" given by former students from the network who are now postdocs or junior PIs. The day concluded with a lunch and poster session. 

2024 Symposium

We held our second PhD and Postdoc Symposium on Friday 4th October , bringing together over 40 PhD students and Postdocs working with faculty from the ELLIS Unit Lisbon to showcase their exciting work in the fields of AI and ML. 

The symposium featured flashtalks by PhD students, a poster session with industry posters  (Feedzai, Priberam, NeuralShift & Zendesk), and a career development panel focusing on the transition from PhD/postdoc to group leader. Our invited keynote was André Cruz, a PhD student of ELLIS Fellow Moritz Hardt working at the Max Planck Institute for Intelligent Systems in Tubingen. André’s work focuses on responsible AI and his talk was titled “Seven Years of Algorithmic Fairness”.

2025 Symposium

The 3rd Annual ELLIS Unit Lisbon PhD & Postdoc Symposium was held on Friday 10th October at Instituto Superior Técnico, bringing together over 50 participants from across the AI and machine learning community for a day of cutting-edge research presentations, career development discussions, and networking.

The event showcased the breadth and depth of ongoing research in the fields, offering a platform for early-career researchers to present their work and engage with their peers, industry leaders, and academic mentors.

Event Highlights

• Research presentations by PhD students Kevin Alcedo, Miguel Belbute, Afonso Certo, and Nuno Gonçalves, each sharing innovative approaches to key AI/ML challenges.

• A keynote address by Zita Marinho (IST & Google DeepMind), titled "Reinforcement Learning in the Era of LLMs".

• A career panel discussion featuring Ana Saraiva Ayash, Joana Lobo Antunes, Mariana Santa-Marta, and Frederico Lyra, who spoke about alternative career pathways, touching on science communication, research funding, and science diplomacy.

• A closing talk by ELLIS Member Sérgio Pequito, "When Control Theory Meets Distributed Optimization: Faster and More Robust Methods", offering insights into the fusion of theory and real-world algorithmic efficiency.

The event also featured a vibrant poster session, with contributions from researchers and industry partners including Feedzai, OutSystems, Priberam, and EleutherAI.

A special congratulations goes to Pedro P. Santos from Francisco Melo’s lab, who was awarded the Best Poster Prize for his work titled "The Number of Trials Matters in Infinite-Horizon General-Utility Markov Decision Processes."