1. The Project#
The Discrete Brain is an educational initiative oriented to explore the interconnection between discrete and continuous maths for inspiring algorithms and even to neuralize them. The long-term objective is to foster a principled development of Artificial Intelligence (AI).
In our opinion, the AI-Engineer (AIE) should master discrete maths, including probability as a bridge to heuristic solvers of NP-hard problems. Herein, in Distrinite Mathematics (Discrete maths for AI) we will set the basis of alogithmic exploration.
Graphs play a fundamental role in this subject. The basic idea is that the human brain is a graph itself, a network of connected neurons with paths, trees and subgraphs created by evolution. As the brain is the support of information flow, graphs can be explored by launching random walks.
Real cases. A key element of the Discrete Brain is to illustrate the working of algorithms on realistic cases such as the characterization of social networks and brain networks. It is not only a matter of obtaining some statistics of these networks, as it happens in network science. We go beyond this: we want to understand their structure, diffuse information to label unknown nodes from known ones, compute distances between nodes, etc.
Future. This is the first subject of the Discrete Brain initiative. It will be followed by Heursitic Search and hopefully by others. We will be delighted to tutorize research projects oriented to build intelligent networks such as Graph-Neural Networks.
Team. The creative team of this project is composed by professors and PhD students with a strong knowledge of the state-of-the-art of structural methods in AI.
Francisco Escolano. Full professor and coordinator of the project and of this subject.
Ahmed Begga. PhD student and expert in Graph-Neural Networks.
Francisco Escolano Ruíz
Ahmed Begga Hachlafi