publication

2024
- universal representation of permutation-invariant functions on vectors and tensors
  puoya tabaghi and yusu wang. to appear in the 35th international conference on algorithmic learning theory (alt).
- an effective neural model for circuit netlist representation
  zhishang luo, truong son hy, puoya tabaghi, michael defferrard, elahe rezaei, ryan carey, rhett davis, rajeev jain, and yusu wang. to appear in artificial intelligence and statistics conference (aistats).
- learning ultrametric trees for optimal transport regressions
  samantha chen, puoya tabaghi, and yusu wang. to appear in the 38th annual aaai conference on artificial intelligence.
- optimal tree metric matching enables phylogenomic branch length reconciliation
  shayesteh arasti, puoya tabaghi (equal contributions), yasamin tabatabaee, and siavash mirarab. to appear in research in computational molecular biology conference (recomb).

2023
- principal component analysis in space forms
  puoya tabaghi, michael khanzadeh, yusu wang, and sivash mirarab. submitted to siam journal on mathematics of data science (simods).

2022
- learning hyperbolic embedding for phylogenetic tree placement and updates
  yueyu jiang, puoya tabaghi, siavash mirarab. biology 2022, 11, 1256.
- hyperaid: denoising in hyperbolic spaces for tree-fitting and hierarchical clustering
  eli chien, puoya tabaghi, olgica milenkovic. proceedings of the 28th acm sigkdd international conference on knowledge discovery & data mining.
- phylogenetic placement problem: a hyperbolic embedding approach
  yueyu jiang, puoya tabaghi, siavash mirarab. the 19th annual satellite conference of recomb on comparative genomics (recomb-cg).

2021
- highly scalable and provably accurate classification in poincare balls
  eli chien, chao pan, puoya tabaghi, and olgica milenkovic. ieee international conference on data mining (icdm).
- geometry of similarity comparisons
  puoya tabaghi, jianhao peng, olgica milenkovic, and ivan dokmanic. arXiv preprint: 2006.09858.
- on procrustes analysis in hyperbolic space
  puoya tabaghi and ivan dokmanic. ieee signal processing letters.
- linear classifiers in product space forms
  puoya tabaghi, chao pan, eli chien, jianhao peng, and olgica milenkovic. submitted to journal of machine learning research learning (jmlr).
2020
- hyperbolic distance matrices
  puoya tabaghi and ivan dokmanic. proceedings of the 26th acm sigkdd international conference on knowledge discovery & data mining.
2019
- kinetic euclidean distance matrices
  puoya tabaghi, ivan dokmanic, and martin vetterli. ieee transactions on signal processing, vol. 68, pp. 452-465.
- on the move: localization with kinetic euclidean distance matrices
  puoya tabaghi, ivan dokmanic, and martin vetterli. ieee international conference on acoustics, speech and signal processing (icassp).
- learning schatten–von neumann operators
  puoya tabaghi, maarten v. de hoop, and ivan dokmanic. arXiv preprint: 1901.10076.
- real polynomial gram matrices without real spectral factors
  puoya tabaghi and ivan dokmanic. arXiv preprint: 1903.04085.
2015
- class-preserving manifold learning for detection and classification
  puoya tabaghi and mahmood r. azimi-sadjadi. international joint conference on neural networks (ijcnn).