Speakers

Maksim Kitsak (45min)

Recording: here

Title: Geometric Representations of Complementarity-Driven Networks

Abstract: Some networks are shaped not only by similarity but also by the complementarity principle.  Examples of complementarity-driven networks include interdisciplinary collaboration networks, networks of interacting proteins, and, possibly, food webs. Existing network embedding methods are not readily applicable to complementarity-driven networks. This talk will show a proper framework for the representations of complementarity-driven networks and demonstrate its efficiency in network reconstruction tasks.

Andrei Buciulea Vlas (20 min)

Recording: here

Title: Learning Graphs and Simplicial Complexes From Data

Abstract: This abstract explores the use of graphs for representing complex information, where the graph's structure is often unknown and learned from data, assuming pairwise node interactions. The talk introduces the problem of graph learning from networked data, emphasizing higher-order interactions in real-world scenarios. It discusses the importance of learning both the graph topology and these higher-order interactions from data. The methodology for modeling data and incorporating higher-order interactions is explained, and the abstract concludes by highlighting the advantages of this approach over existing alternatives.

Sergey Shvydun (20 min)

Recording: here

Title: System Identification for Temporal Networks

Abstract: Modelling temporal networks is an open problem that has attracted researchers from a diverse range of fields. Currently, the existing modelling solutions of time-evolving graphs do not allow us to provide an accurate graph sequence. In this paper, we examine the network dynamics from a system identification perspective. We prove that any periodic graph sequence can be accurately modelled as a linear process. We propose two algorithms, called Subspace Graph Generator (SG-Gen) and Linear Periodic Graph Generator (LPG-gen), for modelling periodic graph sequences and provide their performance on artificial graph sequences.  We further propose a novel model, called Linear Graph Generator (LG-gen), that can be applied to non-periodic graph sequences. Our experiments on artificial and real networks demonstrate that many temporal networks can be accurately approximated by periodic graph sequences.