3 October 2024
New paper: Positional Attention: Out-of-Distribution Generalization and Expressivity for Neural Algorithmic Reasoning. You can read a brief announcement of it here.
4 May 2024
Our paper Analysis of Corrected Graph Convolutions was accepted at NeurIPS 2024.
3 September 2024
US patent on Local Node Embeddings for Heterogeneous Graphs.
1 July 2024
Aseem Baranwal will join the AI residency program at XTY Labs.
1 July 2024
I wrote a blog at Medium on Random Data and Graph Neural Networks.
24 May 2024
New paper: Analysis of Corrected Graph Convolutions.
4 May 2024
Our paper Simulation of Graph Algorithms with Looped Transformers was accepted at ICML 2024.
5 February 2024
New paper: Simulation of Graph Algorithms with Looped Transformers.
16 January 2024
Our paper Local Graph Clustering with Noisy Labels was accepted at ICLR 2024.
5 January 2024
Aseem Baranwal is one of the five students that got the Cheriton Scholarship for Winter 2024. That’s 10k each year for 2 years.
14 October 2023
New paper: Local Graph Clustering with Noisy Labels.
21 September 2023
Our paper Optimality of Message-Passing Architectures for Sparse Graphs was accepted at NeurIPS 2023.
15 Aug 2023
Our paper Graph Attention Retrospective is now live at JMLR.
29 May 2023
Our paper Graph Attention Retrospective was accepted at JMLR.
18 May 2023
New paper: Optimality of Message-Passing Architectures for Sparse Graphs.
9 May 2023
Aseem Baranwal has been awarded the Ontario Graduate Scholarship (OGS).
8 May 2023
Aseem Baranwal has been awarded the President’s Graduate Scholarship (PGS).
24 April 2023
Our paper Weighted flow diffusion for local graph clustering with node attributes: an algorithm and statistical guarantees was accepted at ICML 2023 as Oral & Poster.
16 February 2023
Our paper Flow-Based Algorithms for Improving Clusters: A Unifying Framework, Software, and Performance was published at SIAM Review.
30 January 2023
New paper: Weighted flow diffusion for local graph clustering with node attributes: an algorithm and statistical guarantees.
21 January 2023
Our paper Effects of Graph Convolutions in Multi-layer Networks was accepted at at ICLR 2023 with the tag notable-top-25%.
10 December 2022
Aseem Baranwal is one of the top 20 reviewers at the Learning on Graphs Conference. See the annoucement here
19 October 2022
New paper: On Classification Thresholds for Graph Attention with Edge Features.
3 October 2022
Aseem Baranwal will join Google as a Student Researcher.
3 May 2022
Shenghao Yang has been awarded the Canada Graduate Scholarship from NSERC (105K for 3 years), 1 of 330 awards nationally in Canada.
30 April 2022
Our paper Graph Attention Retrospective won the best paper award at the GroundedML workshop.
21 April 2022
New paper: Effects of Graph Convolutions in Deep Networks.
8 April 2022
Shenghao Yang will join Amazon NY as an intern in July 2022.
8 April 2022
Kimon Fountoulakis will join Google for 5 months from May to October 2022 as a visiting faculty researcher.
3 March 2022
Our paper Graph Attention Retrospective is 1/2 papers that was selected for presentation at the GroundedML workshop.
28 February 2022
New paper: Graph Attention Retrospective.
18 February 2022
Shenghao Yang is one of the three students that got the Cheriton Scholarship for Winter 2022. That’s 10k each year for 2 years.
5 January 2022
Our paper Flow-based Algorithms for Improving Clusters: A Unifying Framework, Software, and Performance was accepted at SIAM Review.
20 December 2021
Shenghao Yang was featured at Mathematics News.
14 October 2021
Aseem Baranwal has been nominated by the CS department for the PhD Google Fellowship.
28 September 2021
Our paper Local Hyper-flow Diffusion was accepted at NeurIPS 2021.
13 August 2021
Our paper Targeted Pandemic Containment Through Identifying Local Contact Network Bottlenecks was accepted at PLOS Computational Biology.
22 July 2021
Shenghao was a finalist for the best student presentation prize at SIAM ACDA. His talk was on Local Hyper-flow Diffusion. You can find the video here.
26 June 2021
Shenghao will present his work on Local Hyper-flow Diffusion at SIAM ACDA.
31 May 2021
Shenghao Yang is one of the 10 fellows to the Borealis AI Global Fellowship (10K CAD).
16 May 2021
Our paper Graph Convolution for Semi-Supervised Classification: Improved Linear Separability and Out-of-Distribution Generalization was accepted at ICML 21.
23 February 2021
Our paper Statistical guarantees for local graph clustering was accepted at JMLR.
17 February 2021
New paper: Local Hyper-flow Diffusion.
16 February 2021
New paper: Graph Convolution for Semi-Supervised Classification: Improved Linear Separability and Out-of-Distribution Generalization.
9 December 2020
Our paper Parallel and Communication Avoiding Least Angle Regression was accepted at SISC.
14 October 2020
Shenghao Yang is one of the three nominees of the CS department at UWaterloo for the IBM PhD Fellowship Award Program.
18. September 2020
Shenghao Yang is presenting his work on Targeted Pandemic Containment Through Identifying Local Contact Network Bottlenecks at NetSci 2020 on Wed 23rd of September at 15:55pm CEST (or 09:55am EDT). Here is the schedule. Here is the paper on arXiv.
18. September 2020
The revised version of our paper on parallel and communication avoiding least angle regression is on arXiv. Joint work with S. Das, J. Demmel, L. Grigori, M. W. Mahoney and S. Yang.
18. September 2020
Welcome — our lab is officially starting!