Pierre Vandergheynst
Full Professor
Works on graph signal processing, geometric machine learning and inverse problems for high-dimensional data.
Pierre Vandergheynst's lab at EPFL works on signal processing & information theory, learning theory & optimization and geometry processing & modeling. 11 papers since 2023, 7 at top venues, mostly ICLR and NeurIPS. 8 PhD students. 44 alumni. Applications close .
At a glance
- papers, 2023–2026
- 3/ year
- at T1 venues
- 64%
- PhD students
- 8
- MS students
- 0
- alumni listed
- 44
Papers by yearfrom DBLP, 2023–2026
- 3 papers2026 · T1 2 · collecting
- 4 papers2025 · T1 4
- 4 papers2024 · T1 1
- 0 papers2023 · T1 0
T1 venuesOtherVenues: ICLR, NeurIPS, ICML
Research focuspaper titles that use each topic's words
- Learning theory & optimization3 papers
- Geometry processing & modeling1 paper
- Signal processing & information theory0 papers
- geometric machine0 papers
- graph signal0 papers
Paper list
From DBLP.
20263 papers
- Trans. Mach. Learn. Res.A Survey on Over-smoothing and Over-squashing: Unified Propagation Perspectives on Graph Neural Networks and Transformers (opens in a new tab)
- ICLR (T1)Beyond Ensembles: Simulating All-Atom Protein Dynamics in a Learned Latent Space (opens in a new tab)
- ICLR (T1)Carré du champ flow matching: better quality-generalisation tradeoff in generative models (opens in a new tab)
20254 papers
- ICML (T1)Boosting Protein Graph Representations through Static-Dynamic Fusion (opens in a new tab)
- NeurIPS (T1)Generative Modeling of Full-Atom Protein Conformations using Latent Diffusion on Graph Embeddings (opens in a new tab)
- NeurIPS (T1)On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learning (opens in a new tab)
- NeurIPS (T1)Return of ChebNet: Understanding and Improving an Overlooked GNN on Long Range Tasks (opens in a new tab)
20244 papers
- MLCBGraph learning for capturing long-range dependencies in protein structures (opens in a new tab)
- ICLR (T1)Implicit Gaussian process representation of vector fields over arbitrary latent manifolds (opens in a new tab)
- Nat. Mac. Intell.Molecular set representation learning (opens in a new tab)
- MLCBTowards improving full-length ribosome density prediction by bridging sequence and graph-based representations (opens in a new tab)
Where alumni went44 listed
Apply to EPFLFall 2027
EDIC Computer and Communication Sciences PhD
- Round
- Round 1
- Opens
- —
- GRE
- optional
- English
- TOEFL 100 / IELTS 7.0
- Fee
- none
Confirmed on the program page.
Similar labsnearest by research summary
- CNSJTU
Dequan Wang
13 T1 - USUCSD
Yu-Xiang Wang
- Reinforcement learning
- CNZhejiang
Jun Xiao
60 T1 - DETUM
Reinhard Heckel
- Signal processing & information theory
- CNSJTU
Ruidan Su
0 T1 - CNTsinghua
Longbo Huang
- Reinforcement learning
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The lab site and the paper record do not agree.
- Collected
- Verified
- Papers
- DBLP
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
EPFLcloses