Statistical Physics of Machine Learning
Scientic Supervisor / Contact Person
Name and Surname
Beatriz Seoane Bartolomé
ORCID (link)
Researcher ID (link)
Localization & Research Area
Faculty / Institute
Faculty of Physical Science
Department
Departamento de Física Teórica
Research Area
Information Science and Engineering (ENG), Mathematics (MAT), Physics (PHY)
MSCA & ERC experience
Research group / research team hosted any MSCA fellow?
No
Research group / research team have any ERC beneficiaries?
No
Research Team & Research Topic
Research Team / Research Group Name (if any)
Dynamics of Disordered Systems
Website of the Research team / Research Group / Department
Brief description of the Research Team / Research Group / Department
Our research group is interested in the study of modern machine learning techniques through the lens of statistical mechanics and disordered systems theory. We are leaders in the study of energy-based generative models and, in particular, in the analysis and use of Boltzmann machines.
We are engaged in a wide range of studies, from the fundamental understanding of how generative models encode patterns during the learning process to the improvement of training protocols and architectures using insights from computational and statistical physics. In addition, we are dedicated to developing a new generation of inference tools that facilitate the extraction of clear, interpretable information from large data sets, with a particular focus on practical biological applications. These efforts also aim to increase the transparency and interpretability of neural networks to ensure their responsible and sustainable use in scientific research.
We are engaged in a wide range of studies, from the fundamental understanding of how generative models encode patterns during the learning process to the improvement of training protocols and architectures using insights from computational and statistical physics. In addition, we are dedicated to developing a new generation of inference tools that facilitate the extraction of clear, interpretable information from large data sets, with a particular focus on practical biological applications. These efforts also aim to increase the transparency and interpretability of neural networks to ensure their responsible and sustainable use in scientific research.
Research lines / projects proposed
- Data-driven applications of Restricted Boltzmann machines (RBMs) in bioinformatics and neuroscience.
- Leveraging RBMs for Efficient Simulations of Physical Systems
- Development of new generative models, training protocols, and interpretative tools.
- Theoretical Analysis and Parameter optimization of simple neural network models.
- Theoretical insights about the effects of limited and/or corrupted data
- Leveraging RBMs for Efficient Simulations of Physical Systems
- Development of new generative models, training protocols, and interpretative tools.
- Theoretical Analysis and Parameter optimization of simple neural network models.
- Theoretical insights about the effects of limited and/or corrupted data
Key words
Application requirements
Professional Experience & Documents
CV, letter of motivation
One Page Proposal
You can attach the 'One Page Proposal' to enhance the attractiveness of your application. Supervisors usually appreciate it. Please take into account your background and the information provided in Research Team & Research Topic section to fill in it.
Submit an application
Forms