Towards an Environment for the Evolution of Artificial General Intelligence

One of the major challenges of AI is the creation of a system capable of self-learning, i.e. Artificial General Intelligence (AGI). This project is one step forward towards AGI.
Master

In this project, a simple environment for the emergence of general intelligence features will be investigated, where agents can evolve through environmental rewards, and learn throughout their lifetime. The chosen control system for agents is spiking neural networks. We want network topologies to be able to evolve and become more complex, i.e. not fixed topology. The weights of the networks cannot be inherited, rather the topology, the type of neurons, and the type of learning are subject to evolution.

Goal

This project involves a combination of the following methods:

  • Artificial spiking neural networks
  • Evolution of neural networks through augmenting topologies
  • Agent-based modelling

Learning outcome

  • Insight into advanced techniques of machine learning
  • Collaboration with researchers in the topic of machine learning, specifically deep learning
  • Possibility to implement and research a novel approach

Qualifications

  • Programming
  • Mathematics
  • Motivation

Supervisors

Collaboration partners

OsloMet