Ameya Velingker (अमेय वेलिंगकर)

Ameya Velingker (अमेय वेलिंगकर)

Research Scientist

Google Research

I am a Senior Research Scientist at Google Research. My research interests are broadly in the area of machine learning and theoretical computer science. I am currently interested in learning on graphs (graph neural networks, graph transformers) and reasoning in ML models, combining tools from algorithms with machine learning. My work has also been used for developing machine learning models for routing in Google Maps. Other areas I have worked in include streaming algorithms, privacy, error-correcting codes, etc.

I received my PhD in Computer Science in 2016 at Carnegie Mellon University, where I was advised by Venkatesan Guruswami and Gary Miller. Afterwards, I was a Research Scientist at École Polytechnique Fédérale de Lausanne (EPFL) from 2016-2018.

In 2011, I completed the Master of Advanced Study in Mathematics at the University of Cambridge under the support of the Gates Cambridge Scholarship. Prior to that, I received an AB in Mathematics and SM in Computer Science (supported by a Siebel Scholars Award) from Harvard University in 2010.

  • Graph learning
  • Reasoning in ML models
  • Theoretical computer science
  • Algorithms


  • Our paper, Low-Width Approximations and Sparsification for Scaling Graph Transformers, has been accepted to the GLFrontiers 2023 workshop at NeurIPS.
  • Our paper, Affinity-Aware Graph Networks, has been accepted to NeurIPS 2023.
  • Our following papers have been accepted to ICML 2023:
    • Exphormer: Sparse Transformers for Graphs
    • Fast (1+ε)-Approximation Algorithms for Binary Matrix Factorization
  • Our paper, Efficient Location Sampling Algorithms for Road Networks, has been accepted to the SODS 2023 workshop at ICML.


(2023). Locality-Aware Graph-Rewiring in GNNs. arXiv:2310.01668.

Cite arXiv

(2023). Affinity-Aware Graph Networks. NeurIPS 2023.

Cite Video arXiv

(2023). Exphormer: Sparse Transformers for Graphs. ICML 2023.

Cite Code Video arXiv URL

(2023). Fast (1+ε)-Approximation Algorithms for Binary Matrix Factorization. ICML 2023.

Cite Video arXiv URL

(2022). Linear space streaming lower bounds for approximating CSPs. STOC 2022.

Cite Video arXiv ECCC URL