“Open Data for Official Statistics, and AI”

Abstract

SPC, The Pacific Community, is the largest intergovernative research organisation in the Pacific. Supported by and serving 27 member states and territories, one of our mission is promoting the adoption of open data and open science principles and tools for decision making. In this talk I will discuss the role of global standards, such as SDMX, in opening Pacific data. I will also showcase our strategy to leverage novel LLMs technologies to improve our user experiences, while not getting lost in the AI hype.

“Complex Networks models beyond the discrete”

Abstract

Complex networks have been, and continue to be, extremely successful to model various natural, economic, social,…, stochastic phenomena where interactions are crucial. Given the success, it is not surprising to find a veritable zoo of statistical models for complex networks, starting from the seminal work of Paul Erdös. In this talk I will defend a number of tenets: almost all complex network models are based on discrete mathematical objects, and have a hard time capturing continuous change process (in time and/or space); even when these models ventured into continuous realms (as in the case of Random Dot Product Graphs), randomness regards links between nodes, and not nodes themselves. Hence, here we put forward some ideas about how to go one step further in making complex networks models fully continuous.