-
datasets to be used in the project's experiments. - Benchmarking of membership inference and model memorization techniques. - Development of gradient-based approximate machine unlearning approaches
-
on relevance, quality and timeliness in the area of the call and alignment with its context in the last five years, by considering: a) Publications; 50% b) Resarch Projects; 30% c) Investigation Track Record. 20
-
: Preference will be given to candidates: With a track record of publications related to task planning for robotics. With training and experience in the use of deep learning and large language models
-
goals: it will enable the periodic collection of data from the same websites, allowing policymakers to track the evolution of the targeted data over time, and it will include a set of heuristics designed