Throughout my academic career, I’ve been fortunate to contribute to a wide variety of research directions, including graph representation learning, computational biology, medical imaging, cross-modal learning, reinforcement learning, explainable AI, and deep learning on low-power/embedded systems. A detailed list of relevant publications is provided below.

Currently, my main research interests are on graph representation learning for algorithmic reasoning. This new and exciting direction seeks to understand and employ the expressive power of GNNs for modelling classical algorithms. The implications are potentially massive: expanding the space of application for classical algorithms, aiding novel discoveries in theoretical computer science, and rigorously benchmarking GNNs. For more details, you may consult these papers, and my talk on the area.

Lately, I have also been highly fascinated by the amazing things that happen when these methods interact synergistically with domain experts. The best example of this is our Nature paper on guiding human intuition with AI.

I would be happy to discuss any project idea at the intersection of the above areas; please check my contact page if you are interested!

Conference Publications

• 

  Affinity-Aware Graph Networks

  by

  Ameya Velingker, Ali Kemal Sinop, Ira Ktena, Petar Veličković and Sreenivas Gollapudi

  in NeurIPS (2023)
• 

  Asynchronous Algorithmic Alignment with Cocycles

  by

  Andrew Dudzik, Tamara von Glehn, Razvan Pascanu and Petar Veličković

  in LoG (Spotlight) (2023)
• 

  Recursive Algorithmic Reasoning

  by

  Jonas Jürß, Dulhan Jayalath and Petar Veličković

  in LoG (Spotlight) (2023)
• 

  Latent Space Representations of Neural Algorithmic Reasoners

  by

  Vladimir V. Mirjanić, Razvan Pascanu and Petar Veličković

  in LoG (2023)
• 

  Parallel Algorithms Align with Neural Execution

  by

  Valerie Engelmayer, Dobrik Georgiev and Petar Veličković

  in LoG (2023)
• 

  Neural Algorithmic Reasoning with Causal Regularisation

  by

  Beatrice Bevilacqua, Kyriacos Nikiforou, Borja Ibarz, Ioana Bica, Michela Paganini, Charles Blundell, Jovana Mitrovic and Petar Veličković

  in ICML (2023)
• 

  Half-Hop: A graph upsampling approach for slowing down message passing

  by

  Mehdi Azabou, Venkataramana Ganesh, Shantanu Thakoor, Chi-Heng Lin, Lakshmi Sathidevi, Ran Liu, Michal Valko, Petar Veličković and Eva L. Dyer

  in ICML (2023)
• 

  Dual Algorithmic Reasoning

  by

  Danilo Numeroso, Davide Bacciu and Petar Veličković

  in ICLR (Spotlight) (2023)
• 

  A Generalist Neural Algorithmic Learner

  by

  Borja Ibarz, Vitaly Kurin, George Papamakarios, Kyriacos Nikiforou, Mehdi Bennani, Róbert Csordás, Andrew Dudzik, Matko Bošnjak, Alex Vitvitskyi, Yulia Rubanova, Andreea Deac, Beatrice Bevilacqua, Yaroslav Ganin, Charles Blundell and Petar Veličković

  in LoG (Spotlight) (2022)
• 

  Expander Graph Propagation

  by

  Andreea Deac, Marc Lackenby and Petar Veličković

  in LoG (2022)
• 

  Reasoning-Modulated Representations

  by

  Petar Veličković, Matko Bošnjak, Thomas Kipf, Alexander Lerchner, Raia Hadsell, Razvan Pascanu and Charles Blundell

  in LoG (2022)
• 

  Continous Neural Algorithmic Planners

  by

  Yu He, Petar Veličković, Pietro Liò and Andreea Deac

  in LoG (2022)
• 

  Learnable Commutative Monoids for Graph Neural Networks

  by

  Euan Ong and Petar Veličković

  in LoG (2022)
• 

  Learning Graph Search Heuristics

  by

  Michal Pándy, Weikang Qiu, Gabriele Corso, Petar Veličković, Rex Ying, Jure Leskovec and Pietro Liò

  in LoG (2022)
• 

  Graph Neural Networks are Dynamic Programmers

  by

  Andrew Dudzik and Petar Veličković

  in NeurIPS (2022)
• 

  Learning to Configure Computer Networks with Neural Algorithmic Reasoning

  by

  Luca Beurer-Kellner, Martin Vechev, Laurent Vanbever and Petar Veličković

  in NeurIPS (2022)
• 

  The CLRS Algorithmic Reasoning Benchmark

  by

  Petar Veličković, Adrià Puigdomènech Badia, David Budden, Razvan Pascanu, Andrea Banino, Misha Dashevskiy, Raia Hadsell and Charles Blundell

  in ICML (2022)
• 

  Utility of Equivariant Message Passing in Cortical Mesh Segmentation

  by

  Dániel Unyi, Ferdinando Insalata, Petar Veličković and Bálint Gyires-Tóth

  in MIUA (2022)
• 

  Large-Scale Representation Learning on Graphs via Bootstrapping

  by

  Shantanu Thakoor, Corentin Tallec, Mohammad Gheshlaghi Azar, Mehdi Azabou, Eva L. Dyer, Rémi Munos, Petar Veličković and Michal Valko

  in ICLR (2022)
• 

  Simple GNN Regularisation for 3D Molecular Property Prediction and Beyond

  by

  Jonathan Godwin, Michael Schaarschmidt, Alexander Gaunt, Alvaro Sanchez-Gonzalez, Yulia Rubanova, Petar Veličković, James Kirkpatrick and Peter Battaglia

  in ICLR (2022)
• 

  Algorithmic Concept-based Explainable Reasoning

  by

  Dobrik Georgiev, Pietro Barbiero, Dmitry Kazhdan, Petar Veličković and Pietro Liò

  in AAAI (2022)
• 

  Neural Algorithmic Reasoners are Implicit Planners

  by

  Andreea Deac, Petar Veličković, Ognjen Milinković, Pierre-Luc Bacon, Jian Tang and Mladen Nikolić

  in NeurIPS (Spotlight) (2021)
• 

  How to transfer algorithmic reasoning knowledge to learn new algorithms?

  by

  Louis-Pascal A. C. Xhonneux, Andreea Deac, Petar Veličković and Jian Tang

  in NeurIPS (2021)
• 

  Neural Distance Embeddings for Biological Sequences

  by

  Gabriele Corso, Rex Ying, Michal Pándy, Petar Veličković, Jure Leskovec and Pietro Liò

  in NeurIPS (2021)
• 

  ETA Prediction with Graph Neural Networks in Google Maps

  by

  Austin Derrow-Pinion, Jennifer She, David Wong, Oliver Lange, Todd Hester, Luis Perez, Marc Nunkesser, Seongjae Lee, Xueying Guo, Brett Wiltshire, Peter W. Battaglia, Vishal Gupta, Ang Li, Zhongwen Xu, Alvaro Sanchez-Gonzalez, Yujia Li and Petar Veličković

  in CIKM (Spotlight) (2021)
• 

  On the role of planning in model-based deep reinforcement learning

  by

  Jessica B. Hamrick, Abram L. Friesen, Feryal Behbahani, Arthur Guez, Fabio Viola, Sims Witherspoon, Thomas Anthony, Lars Buesing, Petar Veličković and Théophane Weber

  in ICLR (2021)
• 

  Pointer Graph Networks

  by

  Petar Veličković, Lars Buesing, Matthew C. Overlan, Razvan Pascanu, Oriol Vinyals and Charles Blundell

  in NeurIPS (Spotlight) (2020)
• 

  Principal Neighbourhood Aggregation for Graph Nets

  by

  Gabriele Corso, Luca Cavalleri, Dominique Beaini, Pietro Liò and Petar Veličković

  in NeurIPS (2020)
• 

  Neural Execution of Graph Algorithms

  by

  Petar Veličković, Rex Ying, Matilde Padovano, Raia Hadsell and Charles Blundell

  in ICLR (2020)
• 

  Deep Graph Infomax

  by

  Petar Veličković, William Fedus, William L. Hamilton, Pietro Liò, Yoshua Bengio and R Devon Hjelm

  in ICLR (2019)
• 

  Convolutional neural networks for mesh-based parcellation of the cerebral cortex

  by

  Guillem Cucurull, Konrad Wagstyl, Arantxa Casanova, Petar Veličković, Estrid Jakobsen, Michal Drozdzal, Adriana Romero, Alan Evans and Yoshua Bengio

  in MIDL (2018)
• 

  Automatic Inference of Cross-Modal Connection Topologies for X-CNNs

  by

  Laurynas Karazija, Petar Veličković and Pietro Liò

  in ISNN (2018)
• 

  Cross-modal Recurrent Models for Weight Objective Prediction from Multimodal Time-series Data

  by

  Petar Veličković, Laurynas Karazija, Nicholas D. Lane, Sourav Bhattacharya, Edgar Liberis, Pietro Liò, Angela Chieh, Otmane Bellahsen and Matthieu Vegreville

  in PervasiveHealth (2018)
• 

  Graph Attention Networks

  by

  Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò and Yoshua Bengio

  in ICLR (2018)
• 

  Multi-omics data integration using cross-modal neural networks

  by

  Ioana Bica, Petar Veličković, Hui Xiao and Pietro Liò

  in ESANN (2018)
• 

  Using deep data augmentation training to address software and hardware heterogeneities in wearable and smartphone sensing devices

  by

  Akhil Mathur, Tianlin Zhang, Sourav Bhattacharya, Petar Veličković, Leonid Joffe, Nicholas D. Lane, Fahim Kawsar and Pietro Liò

  in IPSN (2018)
• 

  Scaling health analytics to millions without compromising privacy using deep distributed behavior models

  by

  Petar Veličković, Nicholas D. Lane, Sourav Bhattacharya, Angela Chieh, Otmane Bellahsen and Matthieu Vegreville

  in PervasiveHealth (2017)
• 

  X-CNN: Cross-modal convolutional neural networks for sparse datasets

  by

  Petar Veličković, Duo Wang, Nicholas D. Lane and Pietro Liò

  in SSCI (2016)
• 

  Viral: Real-world competing process simulations on multiplex networks

  by

  Petar Veličković, Andrej Ivašković, Stella Lau and Miloš Stanojević

  in BelBi (2016)

Journal Publications

• 

  Time-warping invariant quantum recurrent neural networks via quantum-classical adaptive gating

  by

  Ivana Nikoloska, Osvaldo Simeone, Leonardo Banchi and Petar Veličković

  in Machine Learning: Science and Technology (2023)
• 

  Scientific discovery in the age of artificial intelligence

  by

  Hanchen Wang, Tianfan Fu, Yuanqi Du, Wenhao Gao, Kexin Huang, Ziming Liu, Payal Chandak, Shengchao Liu, Peter Van Katwyk, Andreea Deac, Anima Anandkumar, Karianne Bergen, Carla P. Gomes, Shirley Ho, Pushmeet Kohli, Joan Lasenby, Jure Leskovec, Tie-Yan Liu, Arjun Manrai, Debora Marks, Bharath Ramsundar, Le Song, Jimeng Sun, Jian Tang, Petar Veličković, Max Welling, Linfeng Zhang, Connor W. Coley, Yoshua Bengio and Marinka Zitnik

  in Nature (2023)
• 

  Combinatorial Optimization and Reasoning with Graph Neural Networks

  by

  Quentin Cappart, Didier Chételat, Elias B. Khalil, Andrea Lodi, Christopher Morris and Petar Veličković

  in JMLR (2023)
• 

  Everything is connected: Graph neural networks

  by

  Petar Veličković

  in Current Opinion in Structural Biology (2023)
• 

  Towards combinatorial invariance for Kazhdan-Lusztig polynomials

  by

  Charles Blundell, Lars Buesing, Alex Davies, Petar Veličković and Geordie Williamson

  in Representation Theory (2022)
• 

  Advancing mathematics by guiding human intuition with AI

  by

  Alex Davies, Petar Veličković, Lars Buesing, Sam Blackwell, Daniel Zheng, Nenad Tomašev, Richard Tanburn, Peter Battaglia, Charles Blundell, András Juhász, Marc Lackenby, Geordie Williamson, Demis Hassabis and Pushmeet Kohli

  in Nature (2021)
• 

  Neural Algorithmic Reasoning

  by

  Petar Veličković and Charles Blundell

  in Patterns (2021)
• 

  ChronoMID—Cross-modal neural networks for 3-D temporal medical imaging data

  by

  Alexander G. Rakowski, Petar Veličković, Enrico Dall’Ara and Pietro Liò

  in PLOS ONE (2020)
• 

  XFlow: Cross-Modal Deep Neural Networks for Audiovisual Classification

  by

  Cătălina Cangea, Petar Veličković and Pietro Liò

  in TNNLS (2019)
• 

  Attentive cross-modal paratope prediction

  by

  Andreea Deac, Petar Veličković and Pietro Sormanni

  in JCB (2019)
• 

  Parapred: antibody paratope prediction using convolutional and recurrent neural networks

  by

  Edgar Liberis, Petar Veličković, Pietro Sormanni, Michele Vendruscolo and Pietro Liò

  in Bioinformatics (2018)
• 

  Muxstep: an open-source C++ multiplex HMM library for making inferences on multiple data types

  by

  Petar Veličković and Pietro Liò

  in Bioinformatics (2016)
• 

  Molecular multiplex network inference using Gaussian mixture hidden Markov models

  by

  Petar Veličković and Pietro Liò

  in Journal of Complex Networks (2016)

Workshop Publications

• 

  How does over-squashing affect the power of GNNs?

  by

  Francesco Di Giovanni, T. Konstantin Rusch, Michael M. Bronstein, Andreea Deac, Marc Lackenby, Siddhartha Mishra and Petar Veličković

  in NeurIPS-W (GLFrontiers) (Spotlight) (2023)
• 

  Geometric Paratope and Epitope Prediction

  by

  Marco Pegoraro, Clémentine Dominé, Emanuele Rodolà, Petar Veličković and Andreea Deac

  in NeurIPS-W (NeurReps) (2023)
• 

  Cayley Graph Propagation

  by

  JJ Wilson and Petar Veličković

  in NeurIPS-W (NeurReps) (2023)
• 

  Neural Priority Queues for Graph Neural Networks

  by

  Rishabh Jain, Petar Veličković and Pietro Liò

  in ICML-W (KLR) (Spotlight) (2023)
• 

  Sheaf Neural Networks with Connection Laplacians

  by

  Federico Barbero, Cristian Bodnar, Haitz Sáez de Ocáriz Borde, Michael Bronstein, Petar Veličković and Pietro Liò

  in ICML-W (TAG) (2022)
• 

  Message passing all the way up

  by

  Petar Veličković

  in ICLR-W (GTRL) (Spotlight) (2022)
• 

  Learning Heuristics for A*

  by

  Danilo Numeroso, Davide Bacciu and Petar Veličković

  in ICLR-W (GroundedML) (2022)
• 

  Large-scale graph representation learning with very deep GNNs and self-supervision

  by

  Ravichandra Addanki, Peter W. Battaglia, David Budden, Andreea Deac, Jonathan Godwin, Thomas Keck, Wai Lok Sibon Li, Alvaro Sanchez-Gonzalez, Jacklynn Stott, Shantanu Thakoor and Petar Veličković

  in KDD Cup (2021)
• 

  Neural message passing for joint paratope-epitope prediction

  by

  Alice Del Vecchio, Andreea Deac, Pietro Liò and Petar Veličković

  in ICML-W (CompBio) (2021)
• 

  Persistent Message Passing

  by

  Heiko Strathmann, Mohammadamin Barekatain, Charles Blundell and Petar Veličković

  in ICLR-W (SimDL) (2021)
• 

  Predicting Patient Outcomes with Graph Representation Learning

  by

  Catherine Tong, Emma Rocheteau, Petar Veličković, Nicholas Lane and Pietro Liò

  in AAAI-W (W3PHAI) (2021)
• 

  XLVIN: eXecuted Latent Value Iteration Nets

  by

  Andreea Deac, Petar Veličković, Ognjen Milinković, Pierre-Luc Bacon, Jian Tang and Mladen Nikolić

  in NeurIPS-W (Deep RL) (2020)
• 

  A step towards neural genome assembly

  by

  Lovro Vrček, Petar Veličković and Mile Šikić

  in NeurIPS-W (LMCA) (2020)
• 

  Hierarchical Protein Function Prediction with Tail-GNNs

  by

  Stefan Spalević, Petar Veličković, Jovana Kovačević and Mladen Nikolić

  in ICML-W (GRL+) (2020)
• 

  The PlayStation Reinforcement Learning Environment (PSXLE)

  by

  Carlos Purves, Cătălina Cangea and Petar Veličković

  in NeurIPS-W (Deep RL) (2019)
• 

  Drug-Drug Adverse Effect Prediction with Graph Co-Attention

  by

  Andreea Deac, Yu-Hsiang Huang, Petar Veličković, Pietro Liò and Jian Tang

  in ICML-W (CompBio) (2019)
• 

  Spatio-Temporal Deep Graph Infomax

  by

  Felix L. Opolka, Aaron Solomon, Cătălina Cangea, Petar Veličković, Pietro Liò and R Devon Hjelm

  in ICLR-W (RLGM) (2019)
• 

  Towards Sparse Hierarchical Graph Classifiers

  by

  Cătălina Cangea, Petar Veličković, Nikola Jovanović, Thomas Kipf and Pietro Liò

  in NeurIPS-W (R2L) (2018)
• 

  Quantifying the Effects of Enforcing Disentanglement on Variational Autoencoders

  by

  Momchil Peychev, Petar Veličković and Pietro Liò

  in NIPS-W (Disentanglement) (2017)