PhD student at University of Geneva and Hes-so Valais (since 2018)
I am a PhD student with double affiliation at the computer science faculty at the University of Geneva and at the University of Applied Sciences of Western Switzerland (Hes-so Valais). My research aims at improving the interpretability of machine learning systems for healthcare. I was a visiting student at the Martinos Center, part of Harvard Medical School in Boston, MA, USA to focus on the interaction between clinicians and deep learning systems. Coming from a background of IT Engineering, I was awarded the Engineering Department Award for completing the MPhil in Machine Learning, Speech and Language Technology at the University of Cambridge (UK) in 2017.
|Human-centric interpretability for histopathology||Explanations of machine learning models should facilitate the deployment of automated systems in the pathology workflow. To do so, explaining black-box models should take into account the requirements and domain ontology of the end-users.|
|User-based evaluation of explainability methods||Human-centric approaches to explain machine learning decisions consider the user’s need in the XAI development. End-users should be introduced in the evaluation loop to quantify the increase of confidence in the machine learning method, the impact on the final decisions.|
|Interpretability for the media industry||With the AI4Media project, we collaborate to make sure that the European values of ethical and trustworthy AI are embedded in future AI deployments.|
|Epistemic uncertainty weighting of user-controlled losses to guide CNN training||Our recent work on guiding CNN training by user-friendly concepts, e.g. clinical concepts for histopathology, had a main limitation in the formulation of the loss. We address this by epistemic uncertainty weighting.|
During my thesis project, I worked on a spatial smoothing loss function that would allow for target-specific regularization for recurrent natural language models.
I interrupted this MSc to apply to the MPhil in Cambridge, as it was more suited to my academic interests.
Skills in brief I speak Italian, English and French. I understand Spanish and can interact at a basic level. I mostly program in Python, Tensorflow and Keras. I have experience with Matlab and C programming.