Georgios Papagiannis

About

Georgios Papagiannis is a researcher specializing in probabilistic state estimation and machine learning, with a particular focus on the integration of neural networks into sequential Monte Carlo methods. His most notable contribution lies in the development of end-to-end semi-supervised learning frameworks for differentiable particle filters, a cutting-edge approach that bridges classical Bayesian filtering techniques with modern deep learning architectures. By making particle filters differentiable, Papagiannis and his collaborators have enabled the joint learning of both dynamic and measurement models, significantly enhancing the flexibility and scalability of particle filters for large-scale, real-world applications. This work, which has accumulated 15 citations since its 2021 publication, addresses a longstanding challenge in the field: adapting principled probabilistic inference methods to complex, high-dimensional environments where hand-crafted models fall short. His research represents an important step toward more robust and data-driven state estimation systems, with potential applications spanning robotics, autonomous navigation, and tracking. For students and researchers working at the intersection of probabilistic inference and deep learning, Papagiannis's contributions offer a compelling framework for rethinking how classical filtering methods can evolve in the age of neural networks.

Research Focus

Key Achievements

Top Papers

1. 1
   End-to-End Semi-supervised Learning for Differentiable Particle Filters

   15 citations · 2021
2. 2
   End-To-End Semi-supervised Learning for Differentiable Particle Filters

   2 citations · 2020

Key Collaborators

Contact & Links