-
data (sequences and associated properties), adapting existing generative architectures (e.g. VAEs and diffusion-based models when appropriate) to the peptide domain, and defining suitable representations
-
assignment of a research grant, with one position(s), under the project COMPETE2030-FEDER-00819700 | 16914, title W2R - From Waste to Resource: Bioprocesses for recycling mining industry waste in alignment
-
while protecting data privacy. Unlike traditional centralized machine learning, where data must be collected and stored in a central server, FL allows multiple parties to collaboratively build a global
-
. Building on the existing FT-Transformer baseline and SNN-head experiments, the work aims to strengthen empirical evidence, improve interpretability, and align the methodology with Responsible AI principles
-
application of the methodology. By aligning technical innovation with ethics and interpretability, the project aims to contribute to a safer and more responsible adoption of AI in the healthcare domain
-
-Implement a toolkit or pipeline that incorporates multiple curation techniques. 4-Evaluate the impact of each curation step on data quality and model performance using standard ML tasks. 5-Propose a