Networks, or graphs, are a valuable tool in comprehending complex systems by representing a wide range of real-world phenomena such as social networks, biological networks, financial systems, communication networks, and more. Graphs can be used to model many aspects of these real-world phenomena and can help us understand the intricate relationships and interactions within them. Apart from being useful for the study of complex systems, applications of machine learning to graphs have led to important new applications -- from information systems and social media to drug discovery -- and have recently become one of the fastest growing areas in artificial intelligence. However, standard graph representations of complex relational data are limited in their representation capabilities as they only capture dyadic relationships and do not take into account higher-order properties of the system being analyzed.

In recent years, deep learning techniques combined with higher-order network models have gained significant attention as they can effectively capture the multi-relational and multi-dimensional characteristics of complex systems. These models, such as simplicial complexes, manifolds and hypergraph representations, provide a more comprehensive representation of the system, enabling a deeper understanding of the intricate relationships within the data. The utilization of deep learning algorithms in conjunction with such higher-order network models has demonstrated the ability to enhance the identification of previously unobserved characteristics and patterns in complex systems.

Previous editions of HONS have been a valuable platform for researchers to study the challenges and opportunities of higher-order network models. Bridging the network science and deep learning community, this year's edition of HONS will explore how we can integrate higher-order networks into geometric and topological deep learning techniques and how we can use deep learning to facilitate applications of higher-order network models.

The purpose of this event is to create an environment for the exchange of ideas for collaborations between participants and to discuss challenges and opportunities. Some of the topics that will be discussed are:

  • - How can higher-order graph models like simplicial complexes, hypergraphs, memory networks, De Bruijn graphs, etc. best be incorporated into deep learning models?
  • - Which challenges in terms of computational complexity, data efficiency, overfitting does the inclusion of higher-order information entail?
  • - How can we use Graph Neural Networks for an end-to-end higher-order model network selection?
  • - How can we combine statistical inference and model selection for higher-order model into deep learning pipelines?
  • - How do we interpret the result of a higher-order model and what kind of insights they can reveal that traditional models can not?
  • - Which analytical statements can we make about the expressivity of higher-order graph models compared to known results about the expressive power of Graph Neural Networks?
  • - Which additional ethical challenges are introduced by the use of higher-order graph models in deep learning?

We look forward to seeing you at the HONS satellite!

Date and Venue

The satellite will be held on July 10th. This year's satellite will be held in Vienna, Austria.


If you are interested in presenting at the satellite please send us a title and short abstract via E-Mail.