Graph Learning
Graphical models play a viable role in representing complex data in the fields of science and engineering. Social graphs, biological graphs, financial graphs, transportation graphs, sensor graphs and neural graphs are few prominent examples of graph data. When a natural choice of graph is not readily available from the data sets, inferring or learning a graph topology from the data is preferable. MISN is focusing on learning meaningful graph structures derived from artificial and real data sets.
Federated Learning
The development of distributed learning has resulted from the
increasing complexity of machine learning models, as well as the
amount of data required to train them. Federated Learning is a new
paradigm for improving the algorithms that now regulate many areas
of our lives, such as Facebook's News Feed and Google Maps. By
combining knowledge from several data sources, federated learning
has the potential to build extremely accurate statistical models.
Graph Coarsening
Graph coarsening is a widely used dimensionality reduction technique
for approaching large-scale graph machine learning problems. Given a
large graph, graph coarsening aims to learn a smaller-tractable
graph while preserving the properties of the originally given graph.
Graph coarsening is a relatively new approach, and we are developing
graph coarsening algorithms depending on applications such as
neurocomputing, machine learning, big data etc.
Neurocomputing
Graph theory has been around for centuries and has touched several fields as a tool to analyze various complex systems. One such highly complex system is our human brain; hence, graph theory is being used to study the human brain ubiquitously. We are working on a few projects to unfold the mysteries of the human brain using state-of-the-art graph models.
Other Applications
Graph-based machine learning techniques are being increasingly
utilized in climate science to uncover complex relationships and
patterns within the climate system. By representing climate data as
graphs, researchers can gain insights into the organization and
dynamics of climate systems.
