Weijie Su

is this you? claim profile

∙ 0 followers
University of Pennsylvania ∙ Stanford University

Deformable DETR: Deformable Transformers for End-to-End Object Detection

DETR has been recently proposed to eliminate the need for many hand-desi...
10/08/2020 ∙ by Xizhou Zhu, et al. ∙ 10 ∙ share

read it
Benign Overfitting and Noisy Features

Modern machine learning often operates in the regime where the number of...
08/06/2020 ∙ by Zhu Li, et al. ∙ 21 ∙ share

read it
Robust Learning Rate Selection for Stochastic Optimization via Splitting Diagnostic

This paper proposes SplitSGD, a new stochastic optimization algorithm wi...
10/18/2019 ∙ by Matteo Sordello, et al. ∙ 0 ∙ share

read it
VL-BERT: Pre-training of Generic Visual-Linguistic Representations

We introduce a new pre-trainable generic representation for visual-lingu...
08/22/2019 ∙ by Weijie Su, et al. ∙ 8 ∙ share

read it
Algorithmic Analysis and Statistical Estimation of SLOPE via Approximate Message Passing

SLOPE is a relatively new convex optimization procedure for high-dimensi...
07/17/2019 ∙ by Zhiqi Bu, et al. ∙ 0 ∙ share

read it
Quantifying Intrinsic Uncertainty in Classification via Deep Dirichlet Mixture Networks

With the widespread success of deep neural networks in science and techn...
06/11/2019 ∙ by Qingyang Wu, et al. ∙ 0 ∙ share

read it
Statistical Inference for Online Learning and Stochastic Approximation via Hierarchical Incremental Gradient Descent

Stochastic gradient descent (SGD) is an immensely popular approach for o...
02/13/2018 ∙ by Weijie Su, et al. ∙ 0 ∙ share

read it
Statistical Inference for the Population Landscape via Moment Adjusted Stochastic Gradients

Modern statistical inference tasks often require iterative optimization ...
12/20/2017 ∙ by Tengyuan Liang, et al. ∙ 0 ∙ share

read it
When Does the First Spurious Variable Get Selected by Sequential Regression Procedures?

Applied statisticians use sequential regression procedures to produce a ...
08/10/2017 ∙ by Weijie Su, et al. ∙ 0 ∙ share

read it
Private False Discovery Rate Control

We provide the first differentially private algorithms for controlling t...
11/12/2015 ∙ by Cynthia Dwork, et al. ∙ 0 ∙ share

read it
False Discoveries Occur Early on the Lasso Path

In regression settings where explanatory variables have very low correla...
11/05/2015 ∙ by Weijie Su, et al. ∙ 0 ∙ share

read it
Communication-Efficient False Discovery Rate Control via Knockoff Aggregation

The false discovery rate (FDR)---the expected fraction of spurious disco...
06/17/2015 ∙ by Weijie Su, et al. ∙ 0 ∙ share

read it
A Differential Equation for Modeling Nesterov's Accelerated Gradient Method: Theory and Insights

We derive a second-order ordinary differential equation (ODE) which is t...
03/04/2015 ∙ by Weijie Su, et al. ∙ 0 ∙ share

read it

Weijie Su

Featured Co-authors

Xiaogang Wang

Li Zhang

Bin Li

Furu Wei

Zhou Yu

Dino Sejdinovic

Stephen Boyd

Jifeng Dai

Emmanuel J. Candès

Yue Cao

Tengyuan Liang

Deformable DETR: Deformable Transformers for End-to-End Object Detection

Benign Overfitting and Noisy Features

Robust Learning Rate Selection for Stochastic Optimization via Splitting Diagnostic

VL-BERT: Pre-training of Generic Visual-Linguistic Representations

Algorithmic Analysis and Statistical Estimation of SLOPE via Approximate Message Passing

Quantifying Intrinsic Uncertainty in Classification via Deep Dirichlet Mixture Networks

Statistical Inference for Online Learning and Stochastic Approximation via Hierarchical Incremental Gradient Descent

Statistical Inference for the Population Landscape via Moment Adjusted Stochastic Gradients

When Does the First Spurious Variable Get Selected by Sequential Regression Procedures?

Private False Discovery Rate Control

False Discoveries Occur Early on the Lasso Path

Communication-Efficient False Discovery Rate Control via Knockoff Aggregation

A Differential Equation for Modeling Nesterov's Accelerated Gradient Method: Theory and Insights