Roxana Pop, PhD Candidate, is currently investigating neural methods for temporal prediction on structured data, with an emphasis on graph neural networks extensions. Her overarching objective is to combine neural methods with symbolic AI approaches, and extract logical rules from the trained neural networks for making the predictions. In this way, the predictions can be understood by human experts, but the extraction is data-driven, making the best of the two AI paradigms.
Deep learning is at the center of the AI hype we are now living through. And while neural networks are an incredible tool, and have led to great results in many domains, they have their limitations. Most notably, they are black boxes and making neural networks inherently interpretable or providing explanations for their behavior are active fields of research. More traditional AI approaches, the so called symbolic approaches, have complementary benefits to deep learning. Hence, a lot of work has gone into combining the two and constructing hybrid systems. Roxana will discuss one way of doing this combination in the context of spatio-temporal forecasting, based on recent connections made between Graph Neural Networks and logic.
Program
11:30 – Doors open and lunch is served
12:00 – "A Hybrid Approach for Spatio-Temporal Forecasting" by Roxana Pop (PhD Candidate, SIRIUS)
This event is open for all PhD candidates and postdocs. No registration needed.
About the seminar series
Once a month, dScience will invite you to join us for lunch, soft drinks and professional talks at the Science Library. In addition to these, we will serve lunch to PhD candidates in our lounge in Kristine Bonnevies hus every Thursday. Due to limited space (40 people), this will be first come, first served. See how to find us here (download).
Our lounge can also be booked by PhDs and Postdocs on a regular basis, whether it is for a meeting or just to hang out – we have fresh coffee all day long! Read more about the seminar series here.