Research Agenda:
I am currently in the stage of proposing several research trajectories on the context of Applied Deep Learning to Multi-modal data. This is given the current needs of effective means of data processing and analytic to massive amounts of rich, highly contextual multimedia data, with deep learning as tangible solution given its highly flexible and ability to produce accurate inferences on many relevant applications.
Research Trajectories:
There are three research trajectories that I am planning to undergo:
- Multi-modal fusion: this axis focusses on finding the effective means of merging different data modalities by also taking into account underlying data characteristics. Keywords: Geometric Deep Learning, Sensor fusion, Multi-modal learning.
- Feature Learning: this research direction concentrates in discovering automatic learning mechanisms for representative and effecient features generations from input datas. Keywords: Self-supervised learning, Unsupervised Learning, Few shot learnings.
- Models Explainability: this line of research considers the problems of explaining the models predictions, that as such, will be able to instill confident in models predicitons. Keywords: Explainability, Interpretability.
Stay tune on this website for more updates on these research directions!