Boris Knyazev

About me

I’m a Research Scientist at Samsung - SAIT AI Lab. I completed my PhD at the Machine Learning Research Group, University of Guelph and Vector Institute under supervision of Graham Taylor in 2022. My research interests lie at the intersection of graph neural networks (GNNs), optimization and meta-learning with applications to computer vision, language modeling and molecule discovery. In the past I interned at Facebook AI Research (FAIR) working with ‪Adriana Romero and Michal Drozdzal on parameter prediction for neural networks. I also interned at Mila working with Eugene Belilovsky and Aaron Courville on visual compositional generalization. I also interned at SRI International with Mohamed Amer, where I worked on training GNNs on image superpixels. Before starting my PhD, I worked on unsupervised learning and pretraining of neural networks, face, emotion and facial attributes recognition, and video recognition.

Oct 2024: The paper µLO, is accepted as oral at OPT for ML 2024 NeurIPS Workshop, congrats Benjamin and Charles-Étienne!
Sep 2024: New preprint (NiNo) is available arXiv, pdf, code
Feb 2024: 1 paper accepted as oral at ICLR 2024 openreview, arXiv, code, twitter, congrats Miltos!
Oct 2023: 1 paper is accepted at NeurIPS 2023 Workshop on Federated Learning Learning Optimizers for Local SGD
Jun 2023: 2 papers are accepted at ICML 2023 Workshops LLMs for Graphs, Learning to Optimize
Apr 2023: 1 paper accepted at ICML 2023 arXiv, code
Sep 2022: 1 paper accepted at NeurIPS 2022 Track Datasets and Benchmarks (openreview, arXiv, dataset)
Sep 2022: 1 paper accepted at NeurIPS 2022 (arXiv, NeurIPS)
Mar 2022: My PhD thesis has been approved and is available online, LateX source
Jan 2022: 1 paper accepted at ICLR 2022 (openreview, arXiv, code)
Sep 2021: 2 out of 2 papers accepted at NeurIPS 2021
Aug 2021: Selected as an “Outstanding Reviewer” for ICCV 2021 (top 5% student reviewers)
Jul 2021: 2 out of 2 papers accepted at ICCV 2021

2024: ICML, NeurIPS, ICLR, Neural Networks
2023: ICML, NeurIPs, MLG @ KDD, ICLR
2022: CVPR, ICML, ICML Workshop, NeurIPS, Learning on Graphs Conference (LoG), Nature Machine Intelligence
2021: ICCV (Outstanding Reviewer, top 5% student reviewers)
2020: BMVC, ICML Workshop on Graph Representation Learning and Beyond

See the full list at Google Scholar.

Graph Neural Networks for Learning Equivariant Representations of Neural Networks
Miltiadis Kofinas, Boris Knyazev, Yan Zhang, Yunlu Chen, Gertjan J Burghouts, Efstratios Gavves, Cees GM Snoek, David W Zhang
International Conference on Learning Representations (ICLR), 2024 (oral)
openreview, pdf, code

Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models?
Boris Knyazev, Doha Hwang, Simon Lacoste-Julien
International Conference on Machine Learning (ICML), 2023
html, pdf, video, code

Hyper-Representations as Generative Models: Sampling Unseen Neural Network Weights
Konstantin Schürholt, Boris Knyazev, Xavier Giró-i-Nieto, Damian Borth
Advances in Neural Information Processing Systems (NeurIPS), 2022
html, pdf, slides, code

On Evaluation Metrics for Graph Generative Models
Rylee Thompson, Boris Knyazev, Elahe Ghalebi, Jungtaek Kim, Graham W Taylor
International Conference on Learning Representations (ICLR), 2022
html, pdf, openreview, code

Parameter Prediction for Unseen Deep Architectures
Boris Knyazev, Michal Drozdzal, Graham W. Taylor, Adriana Romero-Soriano
Advances in Neural Information Processing Systems (NeurIPS), 2021
html, pdf, reviews, UofG news, Yannic Kilcher’s video, neurips video, code, Colab-predict, Colab-fine-tune, twitter, quantamagazine

Brick-by-Brick: Combinatorial Construction with Deep Reinforcement Learning
Hyunsoo Chung, Jungtaek Kim, Boris Knyazev, Jinhwi Lee, Graham W. Taylor, Jaesik Park, Minsu Cho
Advances in Neural Information Processing Systems (NeurIPS), 2021
html, pdf, reviews, video, code

Context-aware Scene Graph Generation with Seq2Seq Transformers
Yichao Lu, Himanshu Rai, Jason Chang, Boris Knyazev, Shashank Shekhar, Graham W. Taylor, Maksims Volkovs
International Conference on Computer Vision (ICCV), 2021
html, pdf, code

Generative Compositional Augmentations for Scene Graph Prediction
Boris Knyazev, Harm de Vries, Cătălina Cangea, Graham W. Taylor, Aaron Courville, Eugene Belilovsky
International Conference on Computer Vision (ICCV), 2021
html, pdf, ICML Workshop version, ICML workshop video, ICCV code

Graph Density-Aware Losses for Novel Compositions in Scene Graph Generation
Boris Knyazev, Harm de Vries, Cătălina Cangea, Graham W. Taylor, Aaron Courville, Eugene Belilovsky
British Machine Vision Conference (BMVC), 2020
html, pdf, bmvc, code, Data Fest tutorial

Learning Temporal Attention in Dynamic Graphs with Bilinear Interactions
Boris Knyazev*, Carolyn Augusta*, Graham Taylor (*equal contribution)
PLOS ONE, 2021
html, pdf, plos one journal link, code

Understanding Attention and Generalization in Graph Neural Networks
Boris Knyazev, Graham Taylor, Mohamed Amer
Advances in Neural Information Processing Systems (NeurIPS), 2019
html, pdf, neurips, ICLR Workshop version, code, poster, slides

Image Classification with Hierarchical Multigraph Networks
Boris Knyazev, Xiao Lin, Mohamed Amer, Graham Taylor
British Machine Vision Conference (BMVC), 2019
html, pdf, bmvc pdf, code, blog post

Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules
Boris Knyazev, Xiao Lin, Mohamed Amer, Graham Taylor
NeurIPS Workshop on Machine Learning for Molecules and Materials, 2018
html, pdf, code

Leveraging Large Face Recognition Data for Emotion Classification
Boris Knyazev, Roman Shvetsov, Natalia Efremova, Artem Kuharenko
FG Workshop on Large-scale Emotion Recognition and Analysis (LERA), 2018
html, pdf, code, Top-2 in EmotiW 2017 challenge

Recursive Autoconvolution for Unsupervised Learning of Convolutional Neural Networks
Boris Knyazev, Erhardt Barth, Thomas Martinetz
International Joint Conference on Neural Networks (IJCNN), 2017
html, pdf, matlab code, python code, reddit

Last updated: Nov, 2024