LoG 2024 Vancouver Meetup

 Information

The Vancouver local meetup for the Learning on Graphs (LoG) Conference 2024 aims to create an engaging and collaborative environment for researchers, engineers, and practitioners working on machine learning for graphs and geometry. This event will bring together participants from academia and industry across the Greater Vancouver Area to enhance the virtual conference experience with in-person discussions, networking opportunities, and the exchange of innovative ideas.

By fostering meaningful connections, we hope to inspire new collaborations between local researchers and industry professionals, strengthening the links between academic research and practical applications in this growing field. Join us at the Vancouver meetup and contribute to the global conversation on machine learning for graphs at LoG 2024!

Location and Registration

The meetup will be held on November 26, 2024 at the University of British Columbia (UBC), in the MacLeod Building, Room 3038. There is no registration fee, and food and coffee will be provided. Please use the following link to register: Registration Link

Topics

In addition to the core topics of the LoG conference, we welcome work at the intersection of graphs, geometry, and foundation models, including but not limited to:

  • Large Language Models (LLMs)
  • Multi-modal LLMs
  • Applications in computer vision
  • Natural language processing
  • AI for science
  • AI safety
  • Finance

 Schedule

Time Event
8:45am to 9:00am Opening Remark
9:00am to 9:30am Invited Talk 1: Scaling Graph Transformers with Sparse and Sparsified Attention, Danica Sutherland (UBC)
9:30am to 10:00am Invited Talk 2: Trustworthy machine learning for understanding genome regulation, Maxwell W Libbrecht (SFU)
10:00am to 10:30am Invited Talk 3: Playing with Scene Graphs, Leonid Sigal (UBC)
10:30am to 10:45am Coffee Break 1
10:45am to 11:00am Oral Session 1, Talk 1: SymmetricDiffusers: Learning Discrete Diffusion on Finite Symmetric Groups [Nick Zhang]
11:00am to 11:15am Oral Session 1, Talk 2: PAPR: Proximity Attention Point Rendering [Yanshu Zhang]
11:15am to 11:30am Oral Session 1, Talk 3: Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders [Muchen Li]
11:30am to 11:45am Oral Session 1, Talk 4: Rejection Sampling IMLE: Designing Priors for Better Few-Shot Image Synthesis [Chirag Vashist]
11:45am to 2:00pm Lunch + Poster Session
2:00pm to 2:30pm Invited Talk 4: Attend over more with less, Ke Li (SFU)
2:30pm to 3:00pm Invited Talk 5: Some Progress Towards Artificial Intelligence for Operations Research, Mahdi Mostajabdaveh (Huawei Canada)
3:00pm to 3:30pm Invited Talk 6: Attributed Graph Alignment, Lele Wang (UBC)
3:30pm to 3:45pm Coffee Break 2
3:45pm to 4:00pm Oral Session 2, Talk 1: How Compressible Are Graph Transformers [Hamed Shirzad]
4:00pm to 4:15pm Oral Session 2, Talk 2: AoPS Dataset: Leveraging Online Olympiad-Level Math Problems for LLMs Training and Contamination-Resistant Evaluation [Sadegh Mahdavi]
4:15pm to 4:30pm Oral Session 2, Talk 3: Leveraging MoE-based Large Language Model for Zero-Shot Multi-Task Semantic Communication [Sin-Yu Huang]
4:30pm to 4:45pm Oral Session 2, Talk 4: Generative 3D Part Assembly via Part-Whole-Hierarchy Message Passing [Bi'an Du]
4:45pm to 5:00pm Concluding Remark

 Accepted Posters

No. Title Presenter
1 IceFormer: Accelerated Inference with Long-Sequence Transformers on CPUs Yuzhen Mao
2 Efficient Algorithms for Attributed Graph Alignment with Vanishing Edge Correlation Ziao Wang
3 Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders Muchen Li
4 Implicit Geometry of Next-token Prediction: From Language Sparsity Patterns to Model Representations Yize Zhao
5 Fréchet Video Motion Distance: A Metric for Evaluating Motion Consistency in Videos Jiahe Liu
6 How Compressible Are Graph Transformers Hamed Shirzad
7 PAPR: Proximity Attention Point Rendering Yanshu Zhang
8 Rejection Sampling IMLE: Designing Priors for Better Few-Shot Image Synthesis Chirag Vashist
9 MoFlow: One-Step Flow Matching for Human Trajectory Forecasting via Implicit Maximum Likelihood Estimation Distillation Felix Fu
10 From Graph Diffusion to Graph Classification Jia Jun Cheng Xian
11 DARE the Extreme: Revisiting Delta-Parameter Pruning For Fine-Tuned Models Wenlong Deng
12 MVGamba: Unify 3D Content Generation as State Space Sequence Modeling Zike Wu
13 Deep Generative Models of Subgraph Prediction Erfaneh Mahmoudzadeh

 Organizers

Renjie Liao
Renjie Liao

Assistant Professor, UBC

Oliver Schulte
Oliver Schulte

Professor, SFU

Zirui Zhou
Zirui Zhou

Principal Researcher, Huawei Vancouver

 Event Photos

LoG 2024 Vancouver Meetup Group Photo

Participants of the LoG 2024 Vancouver Local Meetup at the University of British Columbia.

 Sponsors

Huawei Logo