APooladian
PhD student at NYU CDS --- interested in high dimensional statistics, convex analysis, and optimal transport.
New York University
Pinned Repositories
1NN_MapEstimator
Implementation of the 1-Nearest-Neighbor estimator for estimating optimal transport maps, as seen in [Plugin Estimation of Smooth Transport Maps](https://arxiv.org/abs/2107.12364) (Manole et al.; 2021).
APooladian
DebiaserBeware
Repository containing Google Colab code for "Debiaser Beware: Pitfalls of Centering Regularized Transport Maps", which was recently accepted to ICML 2022.
FarkasLayers
Repo for Farkas Layers
kernel-sos-ot
logbarrier
LogBarrier adversarial attack
MFVI
ProxLogBarrierAttack
Public repo for the ProxLogBarrier attack
SinkhornBridges
Plug-in estimation of Schrödinger bridges (see here: https://arxiv.org/abs/2408.11686)
TVProjection
Batch-wise implementation of TV-ball projection algorithm; coded in PyTorch
APooladian's Repositories
APooladian/ProxLogBarrierAttack
Public repo for the ProxLogBarrier attack
APooladian/1NN_MapEstimator
Implementation of the 1-Nearest-Neighbor estimator for estimating optimal transport maps, as seen in [Plugin Estimation of Smooth Transport Maps](https://arxiv.org/abs/2107.12364) (Manole et al.; 2021).
APooladian/SinkhornBridges
Plug-in estimation of Schrödinger bridges (see here: https://arxiv.org/abs/2408.11686)
APooladian/DebiaserBeware
Repository containing Google Colab code for "Debiaser Beware: Pitfalls of Centering Regularized Transport Maps", which was recently accepted to ICML 2022.
APooladian/MFVI
APooladian/FarkasLayers
Repo for Farkas Layers
APooladian/TVProjection
Batch-wise implementation of TV-ball projection algorithm; coded in PyTorch
APooladian/APooladian
APooladian/kernel-sos-ot
APooladian/logbarrier
LogBarrier adversarial attack
APooladian/Notes
Collection of personal course notes
APooladian/ott
Optimal Transport tools implemented with the JAX framework, to get auto-diff, parallel and jit-able computations.
APooladian/OWL_AdvAttack
Adversarial attack with respect to the Ordered Weighted L1 (OWL) norm via the ProxLogBarrier framework
APooladian/riemannian-metric-learning-ot