Hassan Ashtiani

David Cheriton School of Computer Science
University of Waterloo
Office: DC3519
mhzokaei at uwaterloo.ca

I study Computer Science as a PhD student at University of Waterloo. I am also a postgraduate affiliate at Vector institute. Broadly speaking, my research interest is machine learning. I am advised by Professor Shai Ben-David.

A Nice Photo of Mine


I am interested in bridging the gap between supervised and unsupervised learning. I am currently focusing on:
  • Interactive Clustering
  • Sample-efficient Distribution Learning
  • Learning methods for high-dimensional data

  • Updates

    • I was selected as a postgraduate affiliate at Vector institute.
    • See my invited talk at Mcgill University about density estimation, December 2017 [Slides]
    • Our paper on learning mixture models was accepted for AAAI'18.
    • I did an internship at Google (Winter 2017) .
    • Watch my talk on "Clustering with same-cluster queries" at NIPS'16.
    • I was awarded the Cheriton Scholarship for 2016-2017.
    • I was invited to give a talk about Clustering with Advice at university of Tehran, December 2015 [Presentation]
    • Check out my guest lectures for the machine learning course at UWaterloo [Lecture 5, Lecture 6]
    • Thanks to Zokaei part of my last name, I was the last name in the list of reviewers for COLT'14 and COLT'15 :-) [2014, 2015]
    • I had an invited talk about ''distribution-specific learnability'' at university of Tehran, December 2013 [Slides]
    • I got Cheriton Scholarship (Type II) for 2012 to 2014


    1. Agnostic Distribution Learning via Compression [paper]
      Hassan Ashtiani, Shai Ben-David, Abbas Mehrabian
      Arxiv Preprint

    2. Sample-Efficient Learning of Mixtures [paper]
      Hassan Ashtiani, Shai Ben-David, Abbas Mehrabian
      AAAI 2018

    3. Online Nearest Neighbor Search in Hamming Space [paper (full version)]
      Sepehr Eghbali, Hassan Ashtiani, Ladan Tahvildari
      ICDM 2017

    4. Clustering with Same-Cluster Queries [paper, slides, video]
      Hassan Ashtiani, Shrinu Kushagra, Shai Ben-David
      NIPS 2016, Oral Presenattion

    5. A Dimension-Independent Generalization Bound for Kernel Supervised Principal Component Analysis [paper]
      Hassan Ashtiani, Ali Ghodsi
      JMLR Workshop and Conference Proceedings, Volume 44: NIPS Workshop on Feature Extraction: Modern Questions and Challenges, 2015

    6. Representation Learning for Clustering: A Statistical Framework [paper, poster, presentation]
      Hassan Ashtiani, Shai Ben-David
      UAI 2015, Oral Presnetation

    7. Bandit-Based Local Feature Subset Selection [paper]
      Hassan Ashtiani, Majid Nili Ahmadabadi, Babak Nadjar Araabi
      Neurocomputing, 2014

    8. Bandit-Based Structure Learning for Bayesian Network Classifiers [paper]
      Sepehr Eghbali, Hassan Ashtiani, Majid Nili Ahmadabadi, Babak Nadjar Araabi
      ICONIP, 2012

    9. Assessment of input variables determination on the SVM model performance using PCA, Gamma test, and forward selection techniques for monthly stream flow prediction [paper]
      Noori, R., A. R. Karbassi, A. Moghaddamnia, D. Han, Hassan Ashtiani, A. Farokhnia, and M. Ghafari Gousheh
      Journal of Hydrology, 2011

    10. A Disaster Invariant Feature for Localization [paper]
      Behdad Soleimani, Hassan Ashtiani, Behrouz Haji Soleimani, Hadi Moradi
      IROS, 2010

    11. An attention-based approach for learning how to fuse decisions of local experts
      Maryam S. Mirian, Majid Nili Ahmadabadi, Babak Nadjar Araabi and Hassan Ashtiani
      NIPS Workshop on Understanding Multiple Kernel Learning Methods, 2009

    12. Graph signature for self-reconfiguration planning of modules with symmetry [paper]
      Masoud Asadpour, Hassan Ashtiani, Alexander Sprowitz, Auke Ijspeert
      IROS, 2009