Postdoc in Generative Machine Learning for Biomedical Data Full-time NEW

The Di Angelantonio-Ieva group in the Health Data Science Centre at Human Technopole is seeking to recruit up to two highly motivated Postdocs in Generative Machine Learning for Biomedical Data. The postholders will focus on developing and applying state-of-the-art generative models (such as VAEs, GANs, and transformer-based architectures) to large-scale biomedical datasets. These models will be used to work with different types of data from the healthcare and biological domains, including genomic profiles, and clinical event sequences. The postholders will develop advanced modeling techniques to create privacy preserving realistic data, and predict disease trajectories, outcomes, and other clinically relevant endpoints. They will also explore and exploit latent spaces to discover meaningful patterns and unlock new insights from complex biomedical data.
The postholders will work in a multidisciplinary environment collaborating with geneticists, molecular epidemiologists, and other data scientists to advance precision medicine research.

Your main tasks and responsibilities:

– Designing and leading analyses that apply state-of-the-art generative machine learning models (e.g., VAEs, GANs, transformer-based models) to large-scale biomedical and biological data, including developing and optimizing models to predict disease progression and create realistic patient profiles;
– Building and optimizing pipelines for pre-processing and integrating biological data sources (clinical event sequences, genomic sequences, disease codes) into unified patient representations and state sequences for predicting disease progression and outcomes;
– Developing advanced generative models to simulate patient health trajectories, including disease progression, based on real-world data, enhancing predictive modeling and allowing for scenario testing in precision medicine;
– Applying transformer-based architectures to model sequences of clinical events and other time-ordered data to predict the future course of diseases;
– Collaborating with epidemiologists, geneticists, and other colleagues in the Centre to develop and implement robust machine learning frameworks and pipelines focused on disease evolution prediction;
– Interpreting results and communicating findings effectively through manuscripts, presentations, and reports, contributing to high-impact publications

Essential Requirements:

– PhD Degree in a relevant scientific field (e.g. computer science, data science, mathematics, engineering, or related);
– Strong understanding of generative models (e.g., VAEs, GANs, transformer-based models), including experience with their application to biomedical and biological data;
– Experience with machine learning frameworks and programming languages (e.g. Python) for handling large-scale text and structured biomedical data;
– Strong quantitative and analytical skills applied to observational or clinical datasets, and familiarity with techniques for representation learning and sequence modeling;
– A track record of authoring scientific publications, with a focus on machine learning methods and applications;
– Fluency in English.

Tagged as: Computer Science, Data Science, Life Sciences, Mathematics