Projects in continual lifelong deep learning Scalable Machine Learning

Description

Various projects in continual learning are available: in the context of both recurrent and feedforward networks, and in supervised and reinforcement learning.

Some examples of possible thesis topics:

  1. Learning to continually learn, similar to (Shawn et al. 2020) but using recurrent networks and more principled (but also biologically plausible) approaches for meta-learning to continually learn.
  2. Understanding catastrophic forgetting in deep networks from a loss landscape view using methods like (Li et al. 2018)
Beaulieu, Shawn, et al. ‘Learning to Continually Learn’. ArXiv:2002.09571 [Cs, Stat], Mar. 2020. arXiv.org, http://arxiv.org/abs/2002.09571.
Li, Hao, et al. ‘Visualizing the Loss Landscape of Neural Nets’. Advances in Neural Information Processing Systems 31, 2018, pp. 6389–99. Neural Information Processing Systems, http://papers.nips.cc/paper/7875-visualizing-the-loss-landscape-of-neural-nets.pdf.

The Institut für Neuroinformatik (INI) is a central research unit of the Ruhr-Universität Bochum. We aim to understand the fundamental principles through which organisms generate behavior and cognition while linked to their environments through sensory systems and while acting in those environments through effector systems. Inspired by our insights into such natural cognitive systems, we seek new solutions to problems of information processing in artificial cognitive systems. We draw from a variety of disciplines that include experimental approaches from psychology and neurophysiology as well as theoretical approaches from physics, mathematics, electrical engineering and applied computer science, in particular machine learning, artificial intelligence, and computer vision.

Universitätsstr. 150, Building NB, Room 3/32
D-44801 Bochum, Germany

Tel: (+49) 234 32-28967
Fax: (+49) 234 32-14210