Oren Mangoubi

activity

20152025

most citedRapid Mixing of Hamiltonian Monte Carlo on Strongly Log-Concave Distributions

70 citations · 138 across the 21 of their papers we have counts for

collaborators

22 papers

cs.LG2025

Efficient Diffusion Models for Symmetric Manifolds

Oren Mangoubi, Neil He, Nisheeth K. Vishnoi

We introduce a framework for designing efficient diffusion models for $d$ -dimensional symmetric-space Riemannian manifolds, including the torus, sphere, special orthogonal group an…

cs.DS2025

Private Low-Rank Approximation for Covariance Matrices, Dyson Brownian Motion, and Eigenvalue-Gap Bounds for Gaussian Perturbations

Oren Mangoubi, Nisheeth K. Vishnoi

We consider the problem of approximating a $d \times d$ covariance matrix $M$ with a rank- $k$ matrix under $(ε, δ)$ -differential privacy. We present and analyze a complex…

cs.DS2024

Faster Sampling from Log-Concave Densities over Polytopes via Efficient Linear Solvers

Oren Mangoubi, Nisheeth K. Vishnoi

We consider the problem of sampling from a log-concave distribution $π (θ) \propto e^{- f (θ)}$ constrained to a polytope $K := {θ \in R^{d} : A θ \leq b}$ , where $A\in \mathbb{R}^…

math.ST2024★ 1 cited

Singular Subspace Perturbation Bounds via Rectangular Random Matrix Diffusions

Peiyao Lai, Oren Mangoubi

Given a matrix $A \in R^{m \times d}$ with singular values $σ_{1} \geq \dots \geq σ_{d}$ , and a random matrix $G \in R^{m \times d}$ with iid $N (0, T)$ entries for some…

cs.DS2023

Private Covariance Approximation and Eigenvalue-Gap Bounds for Complex Gaussian Perturbations

Oren Mangoubi, Nisheeth K. Vishnoi

We consider the problem of approximating a $d \times d$ covariance matrix $M$ with a rank- $k$ matrix under $(ε, δ)$ -differential privacy. We present and analyze a complex…

cs.DS2022★ 1 cited

Re-Analyze Gauss: Bounds for Private Matrix Approximation via Dyson Brownian Motion

Oren Mangoubi, Nisheeth K. Vishnoi

Given a symmetric matrix $M$ and a vector $λ$ , we present new bounds on the Frobenius-distance utility of the Gaussian mechanism for approximating $M$ by a matrix whose spectrum is…

cs.DS2022

Private Matrix Approximation and Geometry of Unitary Orbits

Oren Mangoubi, Yikai Wu, Satyen Kale +2

Consider the following optimization problem: Given $n \times n$ matrices $A$ and $Λ$ , maximize $⟨ A, U Λ U^{*} ⟩$ where $U$ varies over the unitary group $U (n)$ . Th…

cs.DS2022

Sampling from Log-Concave Distributions over Polytopes via a Soft-Threshold Dikin Walk

Oren Mangoubi, Nisheeth K. Vishnoi

Given a Lipschitz or smooth convex function $f : K \to R$ for a bounded polytope $K \subseteq R^{d}$ defined by $m$ inequalities, we consider the problem of sampli…

cs.DS2021

Sampling from Log-Concave Distributions with Infinity-Distance Guarantees

Oren Mangoubi, Nisheeth K. Vishnoi

For a $d$ -dimensional log-concave distribution $π (θ) \propto e^{- f (θ)}$ constrained to a convex body $K$ , the problem of outputting samples from a distribution $ν$ which is $\varep…

cs.DC2021★ 2 cited

Sync-Switch: Hybrid Parameter Synchronization for Distributed Deep Learning

Shijian Li, Oren Mangoubi, Lijie Xu +1

Stochastic Gradient Descent (SGD) has become the de facto way to train deep neural networks in distributed clusters. A critical factor in determining the training throughput and mo…

cs.LG2020

A Convergent and Dimension-Independent Min-Max Optimization Algorithm

Vijay Keswani, Oren Mangoubi, Sushant Sachdeva +1

We study a variant of a recently introduced min-max optimization framework where the max-player is constrained to update its parameters in a greedy manner until it reaches a first-…

cs.DS2020

Greedy Adversarial Equilibrium: An Efficient Alternative to Nonconvex-Nonconcave Min-Max Optimization

Oren Mangoubi, Nisheeth K. Vishnoi

Min-max optimization of an objective function $f : R^{d} \times R^{d} \to R$ is an important model for robustness in an adversarial setting, with appl…

cs.DS2019★ 4 cited

Faster polytope rounding, sampling, and volume computation via a sublinear "Ball Walk"

Oren Mangoubi, Nisheeth K. Vishnoi

We study the problem of "isotropically rounding" a polytope $K \subset R^{n}$ , that is, computing a linear transformation which makes the uniform distribution on the polytope…

cs.DS2019★ 7 cited

Nonconvex sampling with the Metropolis-adjusted Langevin algorithm

Oren Mangoubi, Nisheeth K. Vishnoi

The Langevin Markov chain algorithms are widely deployed methods to sample from distributions in challenging high-dimensional and non-convex statistics and machine learning applica…

cs.LG2019★ 3 cited

Online Sampling from Log-Concave Distributions

Holden Lee, Oren Mangoubi, Nisheeth K. Vishnoi

Given a sequence of convex functions $f_{0}, f_{1}, \dots, f_{T}$ , we study the problem of sampling from the Gibbs distribution $π_{t} \propto e^{- \sum_{k = 0}^{t} f_{k}}$ for each epoch $t$ in…

math.PR2018★ 4 cited

Simple Conditions for Metastability of Continuous Markov Chains

Oren Mangoubi, Natesh S. Pillai, Aaron Smith

A family ${Q_{β}}_{β \geq 0}$ of Markov chains is said to exhibit $metastable mixing$ with $modes$ $S_{β}^{(1)}, \dots, S_{β}^{(k)}$ if its spectral gap (or some other…

math.PR2018★ 17 cited

Does Hamiltonian Monte Carlo mix faster than a random walk on multimodal densities?

Oren Mangoubi, Natesh S. Pillai, Aaron Smith

Hamiltonian Monte Carlo (HMC) is a very popular and generic collection of Markov chain Monte Carlo (MCMC) algorithms. One explanation for the popularity of HMC algorithms is their…

cs.DS2018★ 25 cited

Dimensionally Tight Bounds for Second-Order Hamiltonian Monte Carlo

Oren Mangoubi, Nisheeth K. Vishnoi

Hamiltonian Monte Carlo (HMC) is a widely deployed method to sample from high-dimensional distributions in Statistics and Machine learning. HMC is known to run very efficiently in…

cs.DS2017★ 2 cited

Convex Optimization with Unbounded Nonconvex Oracles using Simulated Annealing

Oren Mangoubi, Nisheeth K. Vishnoi

We consider the problem of minimizing a convex objective function $F$ when one can only evaluate its noisy approximation $F^$ . Unless one assumes some structure on the noise,…

math.PR2017★ 70 cited

Rapid Mixing of Hamiltonian Monte Carlo on Strongly Log-Concave Distributions

Oren Mangoubi, Aaron Smith

We obtain several quantitative bounds on the mixing properties of the Hamiltonian Monte Carlo (HMC) algorithm for a strongly log-concave target distribution $π$ on $R^{d}$ …

math.PR2016

Rapid Mixing of Geodesic Walks on Manifolds with Positive Curvature

Oren Mangoubi, Aaron Smith

We introduce a Markov chain for sampling from the uniform distribution on a Riemannian manifold $M$ , which we call the $geodesic walk$ . We prove that the mixing…

math.PR2015★ 2 cited

Integral geometry for Markov chain Monte Carlo: overcoming the curse of search-subspace dimensionality

Oren Mangoubi, Alan Edelman

We introduce a method that uses the Cauchy-Crofton formula and a new curvature formula from integral geometry to reweight the sampling probabilities of Metropolis-within-Gibbs algo…