I'm a Senior Lecturer (=Assistant Professor) at the department of statistics and operations research, school of mathematical sciences, Tel Aviv University. Prior to that, I was a postdoctoral research associate at Princeton University's Program in Applied and Computational Mathematics, working in Amit Singer's group. I did my Ph.D. at the department of Computer Science and Applied Mathematics at the Weizmann Institute of Science in Israel, where Boaz Nadler was my doctoral advisor.
My research interests are broadly in the development of methodology for statistics and machine learning. More specifically, my current focus is developing tools for mapping and analyzing volumetric shape spaces. This is motivated by a key challenge in structural biology: the analysis of flexible proteins and other macromolecules from cryoelectron microscopy data sets.
RECRUITING! Looking for PhD/MSc students with a strong mathematical foundation. If you find my research interesting, please email me.
Video Presentations:
• Broad overview of my research at INRIA's DataShape seminar:
Nonparametric estimation of highdimensional shape spaces with applications to structural biology
• Presentation for 3rd year math students in Tel Aviv University (hebrew):
Shape spaces, dimensionality reduction and product manifold factorization
Academic Collaborators:
Joakim AndÃ©n (KTH Royal Institute of Technology)
Alberto Bartesaghi (Duke University)
Tamir Bendory (Tel Aviv University)
Amit Halevi (Princeton University)
Ariel Jaffe (Hebrew University of Jerusalem)
Joe Kileel (University of Texas at Austin)
Boaz Nadler (Weizmann Institute of Science)
Saharon Rosset (Tel Aviv University)
Amit Singer (Princeton University)
Sharon Zhang (Stanford University)
Ye Zhou (Duke University)
Address: Schreiber 202, School of Mathematical Sciences, Tel Aviv University.
Amit Moscovich, Saharon Rosset.
On the crossvalidation bias due to unsupervised preprocessing.
Journal of the Royal Statistical Society Series B (2022).
[
paper

code
]
Ye Zhou, Amit Moscovich, Alberto Bartesaghi.
Datadriven determination of number of discrete conformations in singleparticle cryoEM.
Computer Methods and Programs in Biomedicine (2022).
[
paper
]
Joe Kileel, Amit Moscovich, Nathan Zelesko, Amit Singer.
Manifold learning with arbitrary norms.
Journal of Fourier Analysis and Applications (2021).
[
paper

code

tweet
]
Sharon Zhang, Amit Moscovich, Amit Singer.
Product Manifold Learning.
AISTATS 2021.
[
paper

code

tweet

blogpost
]
Rohan Rao, Amit Moscovich, Amit Singer.
Wasserstein KMeans for Clustering Tomographic Projections.
Machine Learning for Structural Biology workshop, NeurIPS (2020).
[
paper

code

video
]
Amit Moscovich.
Fast calculation of pvalues for onesided KolmogorovSmirnov type statistics.
Submitted.
[
preprint

code

tweet
]
Yaniv Tenzer, Amit Moscovich, MaryFrances Dorn, Boaz Nadler, Clifford Spiegelman.
Beyond trees: Classification with sparse pairwise dependencies.
Journal of Machine Learning Research (JMLR), 2020.
[
paper
]
Nathan Zelesko, Amit Moscovich, Joe Kileel, Amit Singer.
Earthmoverbased manifold learning for analyzing molecular conformation spaces.
IEEE 17th International Symposium on Biomedical Imaging (ISBI), 2020.
[
paper

code
]
Ye Zhou, Amit Moscovich, Tamir Bendory, Alberto Bartesaghi.
Unsupervised particle sorting for highresolution singleparticle cryoEM.
Inverse Problems, 2020.
[
paper
]
Amit Moscovich, Amit Halevi, Joakim Andén, Amit Singer.
CryoEM reconstruction of continuous heterogeneity by Laplacian spectral volumes.
Inverse Problems, 2020.
[
paper

code
]
Amit Moscovich, Ariel Jaffe, Boaz Nadler.
Minimaxoptimal semisupervised regression on unknown manifolds.
Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS), 2017.
[
paper

supplementary

poster

slides

code
]
Amit Moscovich, Boaz Nadler.
Fast calculation of boundary crossing probabilities for Poisson processes.
Statistics & Probability Letters, 2017.
[
paper

poster
]
Amit Moscovich, Boaz Nadler, Clifford Spiegelman.
On the exact BerkJones statistics and their pvalue calculation.
Electronic Journal of Statistics, 2016.
[
paper

supplementary

code

Python

R
]