Her research interests include Machine Learning, Representation Learning, Deep Learning and Relational Reasoning. More specifically, Ines is interested in designing models that can learn representations for complex relational structures such as graphs. Ines is particularly excited about understanding how non-Euclidean geometries (e.g., hyperbolic geometry), can lead to more expressive representations for some types of relational structures.
Ines is also excited by applications in the field of Computer Vision and Natural Language Processing, such as understanding how objects interact in images or how entities are related in Knowledge Graphs. During her studies, she had the chance to work on Question Answering at Microsoft AI and Research in 2017, and also spent the Summer of 2018 at Google Research, where she worked on graph-based Semi-Supervised Learning.