Graph Neural Networks
graph neural networks
Michalis Vazirgiannis/Johannes Lutzeyer
POSITION 1, "Graph Representation Learning with Biomedical Applications": The use of Artificial Intelligence (AI) methodology is currently accelerating progress in the area of drug discovery at an impressive speed. Recent successes include the discovery of antibiotics using AI pipelines (Stokes et al., 2020; Liu et al. 2023) as well as the release of the already very impactful AlphaFold model which predicts the three dimensional structure of proteins (Jumper et al., 2021). This rapid scientific progress is also triggering increased industrial interest with Google’s Deepmind announcing the foundation of a new Alphabet subsidiary called Isomorphic Labs with the goal of industrialising AI-driven drug discovery. We are looking for a candidate willing to work in this exciting and dynamic space of scientific progress. Specifically, we would aim to involve the candidate in several projects in which we explore the potential of Graph Representation Learning methodology in the context of Biomedical applications. POSITION 2, "Multimodal Graph Generative Models": Graph generative models are recently gaining significant interest in current application domains. They are commonly used to model social networks, knowledge graphs, and protein-protein interaction networks. The research to be conducted during this project will capitalize on the potential of graph generative models and recent relevant efforts in the Biomedical domain. We will investigate the challenges of multi modality in the context of defining architectures for graph generation under the proper prompt. We expect our designed architectures to be useful in different areas including power grid/telecom/social networks design.
Steven Schockaert
We are looking for a PhD student to work on explainable Graph Neural Networks for problems in computational chemistry. This is a funded position for 3.5 years at Cardiff University's School of Computer Science & Informatics (UK).
Tarek Besold
At Sony AI, we are searching for a (Senior) Research Scientist Data Mining/Knowledge Discovery & ML to join one of our offices in Barcelona (preferred), Zurich or Tokyo. The role involves working with a highly diverse, international team of scientists and engineers pushing the boundaries of AI/ML research.
Prof. Joschka Boedecker
Full-time PhD positions on planning and learning for automated driving at the Neurorobotics Lab, University of Freiburg, Germany. The project involves working in a team with excellent peers in a larger project with an industry partner.
N/A
Applications are invited for a Postdoctoral Research Associate in the Cardiff University School of Computer Science & Informatics, to work on the EPSRC Open Fellowship project ReStoRe (Reasoning about Structured Story Representations), which is focused on story-level language understanding. The overall aim of this project is to develop methods for learning graph-structured representations of stories. For this post, the specific focus will be on developing neuro-symbolic reasoning strategies to fill the gap between what is explicitly stated in a story and what a human reader would infer by “reading between the lines”.
Francesco Piccialli
Exciting opportunity for early-stage researchers to join the TUAI (Towards an Understanding of Artificial Intelligence) project, a Marie Skłodowska-Curie Doctoral Network funded by the European Union’s Horizon Europe program. We are currently offering PhD positions aimed at fostering transparent, open, and explainable AI through innovative research. The TUAI project aims to bridge technical advancements in AI with societal needs, promoting ethical, responsible, and inclusive AI systems.
N/A
Applications are invited for a Postdoctoral Research Associate post in the Cardiff University School of Computer Science & Informatics, to work on the InteGraL project (“Interpretable Graph-Based Machine Learning”). This Leverhulme Trust funded project is focused on developing alternatives to Graph Neural Networks (GNNs). Its central aim will be to introduce methods that are interpretable by design, while at the same time being more robust than GNNs and generalising better to problem instances that are out of the training set distribution. The models will be applied to problems in Natural Language Processing and Computational Chemistry, among others. More details about the post and instructions on how to apply are available at https://www.jobs.ac.uk/job/DMZ697/research-associate
Mental Simulation, Imagination, and Model-Based Deep RL
Mental simulation—the capacity to imagine what will or what could be—is a salient feature of human cognition, playing a key role in a wide range of cognitive abilities. In artificial intelligence, the last few years have seen the development of methods which are analogous to mental models and mental simulation. In this talk, I will discuss recent methods in deep learning for constructing such models from data and learning to use them via reinforcement learning, and compare such approaches to human mental simulation. While a number of challenges remain in matching the capacity of human mental simulation, I will highlight some recent progress on developing more compositional and efficient model-based algorithms through the use of graph neural networks and tree search.
Spatio-temporal Graph Neural Networks for Motor Imagery EEG Classification
Neuromatch 5