Hello, and welcome to one of my homes on the web! I’m Petar, a Senior Research Scientist at DeepMind. I have a PhD in Computer Science from the University of Cambridge (Trinity College), obtained under the supervision of Pietro Liò. My research concerns geometric deep learning—devising neural network architectures that respect the invariances and symmetries in data (a topic I’ve co-written a proto-book about). Within this area, I focus on graph representation learning and its applications in algorithmic reasoning and computational biology. I have published relevant research in these areas at both machine learning venues (NeurIPS, ICLR, ICML-W) and biomedical venues and journals (Bioinformatics, PLOS One, JCB, PervasiveHealth). In particular, I am the first author of Graph Attention Networks—a popular convolutional layer for graphs—and Deep Graph Infomax—a scalable local/global unsupervised learning pipeline for graphs (featured in ZDNet). Further, my research has been used in substantially improving the travel-time predictions in Google Maps (covered by outlets including the CNBC, Endgadget, VentureBeat, CNET, the Verge and ZDNet).
Our Opinion paper on Neural Algorithmic Reasoninghas been accepted for publication in Patterns!
Our proto-book on Geometric Deep Learningis out now on the arXiv!
I gave a talk on the Theoretical Foundations of Graph Neural Networksat the University of Cambridge Computer Laboratory's flagship seminar series!