Search Results for author: Ju Sun

Found 27 papers, 10 papers with code

Practical Phase Retrieval Using Double Deep Image Priors

no code implementations2 Nov 2022 Zhong Zhuang, David Yang, Felix Hofmann, David Barmherzig, Ju Sun

Phase retrieval (PR) concerns the recovery of complex phases from complex magnitudes.

Retrieval

Imbalanced Classification in Medical Imaging via Regrouping

no code implementations21 Oct 2022 Le Peng, Yash Travadi, Rui Zhang, Ying Cui, Ju Sun

We propose performing imbalanced classification by regrouping majority classes into small classes so that we turn the problem into balanced multiclass classification.

Image Classification imbalanced classification +1

NCVX: A General-Purpose Optimization Solver for Constrained Machine and Deep Learning

no code implementations3 Oct 2022 Buyun Liang, Tim Mitchell, Ju Sun

Imposing explicit constraints is relatively new but increasingly pressing in deep learning, stimulated by, e. g., trustworthy AI that performs robust optimization over complicated perturbation sets and scientific applications that need to respect physical laws and constraints.

Optimization for Robustness Evaluation beyond $\ell_p$ Metrics

no code implementations2 Oct 2022 Hengyue Liang, Buyun Liang, Ying Cui, Tim Mitchell, Ju Sun

Empirical evaluation of deep learning models against adversarial attacks entails solving nontrivial constrained optimization problems.

Early Stopping for Deep Image Prior

1 code implementation11 Dec 2021 Hengkang Wang, Taihui Li, Zhong Zhuang, Tiancong Chen, Hengyue Liang, Ju Sun

In this regard, the majority of DIP works for vision tasks only demonstrates the potential of the models -- reporting the peak performance against the ground truth, but provides no clue about how to operationally obtain near-peak performance without access to the groundtruth.

NCVX: A User-Friendly and Scalable Package for Nonconvex Optimization in Machine Learning

1 code implementation27 Nov 2021 Buyun Liang, Tim Mitchell, Ju Sun

GRANSO is among the first optimization solvers targeting general nonsmooth NCVX problems with nonsmooth constraints, but, as it is implemented in MATLAB and requires the user to provide analytical gradients, GRANSO is often not a convenient choice in machine learning (especially deep learning) applications.

BIG-bench Machine Learning

Self-Validation: Early Stopping for Single-Instance Deep Generative Priors

2 code implementations23 Oct 2021 Taihui Li, Zhong Zhuang, Hengyue Liang, Le Peng, Hengkang Wang, Ju Sun

Recent works have shown the surprising effectiveness of deep generative models in solving numerous image reconstruction (IR) tasks, even without training data.

Image Reconstruction

Phase Retrieval using Single-Instance Deep Generative Prior

no code implementations9 Jun 2021 Kshitij Tayal, Raunak Manekar, Zhong Zhuang, David Yang, Vipin Kumar, Felix Hofmann, Ju Sun

Several deep learning methods for phase retrieval exist, but most of them fail on realistic data without precise support information.

Retrieval

Rethinking Transfer Learning for Medical Image Classification

1 code implementation9 Jun 2021 Le Peng, Hengyue Liang, Gaoxiang Luo, Taihui Li, Ju Sun

For example, on the BIMCV COVID-19 classification dataset, we obtain improved performance with around $1/4$ model size and $2/3$ inference time compared to the standard full TL model.

Image Classification Medical Image Classification +1

Towards Low-Photon Nanoscale Imaging: Holographic Phase Retrieval via Maximum Likelihood Optimization

no code implementations24 May 2021 David A. Barmherzig, Ju Sun

A new algorithmic framework is presented for holographic phase retrieval via maximum likelihood optimization, which allows for practical and robust image reconstruction.

Image Reconstruction Retrieval

Deep Learning Initialized Phase Retrieval

no code implementations23 Oct 2020 Raunak Manekar, Zhong Zhuang, Kshitij Tayal, Vipin Kumar, Ju Sun

Phase retrieval (PR) consists of estimating 2D or 3D objects from their Fourier magnitudes and takes a central place in scientific imaging.

Retrieval

Inverse Problems, Deep Learning, and Symmetry Breaking

no code implementations20 Mar 2020 Kshitij Tayal, Chieh-Hsin Lai, Vipin Kumar, Ju Sun

In many physical systems, inputs related by intrinsic system symmetries are mapped to the same output.

Retrieval

Subgradient Descent Learns Orthogonal Dictionaries

1 code implementation ICLR 2019 Yu Bai, Qijia Jiang, Ju Sun

This paper concerns dictionary learning, i. e., sparse coding, a fundamental representation learning problem.

Dictionary Learning Representation Learning

Weighted AdaGrad with Unified Momentum

no code implementations10 Aug 2018 Fangyu Zou, Li Shen, Zequn Jie, Ju Sun, Wei Liu

Integrating adaptive learning rate and momentum techniques into SGD leads to a large class of efficiently accelerated adaptive stochastic algorithms, such as Nadam, AccAdaGrad, \textit{etc}.

Stochastic Optimization

A Local Analysis of Block Coordinate Descent for Gaussian Phase Retrieval

no code implementations6 Dec 2017 David Barmherzig, Ju Sun

While convergence of the Alternating Direction Method of Multipliers (ADMM) on convex problems is well studied, convergence on nonconvex problems is only partially understood.

Retrieval

A Geometric Analysis of Phase Retrieval

1 code implementation22 Feb 2016 Ju Sun, Qing Qu, John Wright

complex Gaussian) and the number of measurements is large enough ($m \ge C n \log^3 n$), with high probability, a natural least-squares formulation for GPR has the following benign geometric structure: (1) there are no spurious local minimizers, and all global minimizers are equal to the target signal $\mathbf x$, up to a global phase; and (2) the objective function has a negative curvature around each saddle point.

GPR Retrieval

Complete Dictionary Recovery over the Sphere II: Recovery by Riemannian Trust-region Method

no code implementations15 Nov 2015 Ju Sun, Qing Qu, John Wright

We consider the problem of recovering a complete (i. e., square and invertible) matrix $\mathbf A_0$, from $\mathbf Y \in \mathbb{R}^{n \times p}$ with $\mathbf Y = \mathbf A_0 \mathbf X_0$, provided $\mathbf X_0$ is sufficiently sparse.

Dictionary Learning

Complete Dictionary Recovery over the Sphere I: Overview and the Geometric Picture

no code implementations11 Nov 2015 Ju Sun, Qing Qu, John Wright

We give the first efficient algorithm that provably recovers $\mathbf A_0$ when $\mathbf X_0$ has $O(n)$ nonzeros per column, under suitable probability model for $\mathbf X_0$.

Dictionary Learning

When Are Nonconvex Problems Not Scary?

3 code implementations21 Oct 2015 Ju Sun, Qing Qu, John Wright

In this note, we focus on smooth nonconvex optimization problems that obey: (1) all local minimizers are also global; and (2) around any saddle point or local maximizer, the objective has a negative directional curvature.

Dictionary Learning Retrieval +1

Complete Dictionary Recovery over the Sphere

1 code implementation26 Apr 2015 Ju Sun, Qing Qu, John Wright

We consider the problem of recovering a complete (i. e., square and invertible) matrix $\mathbf A_0$, from $\mathbf Y \in \mathbb R^{n \times p}$ with $\mathbf Y = \mathbf A_0 \mathbf X_0$, provided $\mathbf X_0$ is sufficiently sparse.

Dictionary Learning

Finding a sparse vector in a subspace: Linear sparsity using alternating directions

1 code implementation NeurIPS 2014 Qing Qu, Ju Sun, John Wright

In this paper, we focus on a **planted sparse model** for the subspace: the target sparse vector is embedded in an otherwise random subspace.

Dictionary Learning

Efficient Point-to-Subspace Query in $\ell^1$ with Application to Robust Object Instance Recognition

no code implementations2 Aug 2012 Ju Sun, Yuqian Zhang, John Wright

Motivated by vision tasks such as robust face and object recognition, we consider the following general problem: given a collection of low-dimensional linear subspaces in a high-dimensional ambient (image) space, and a query point (image), efficiently determine the nearest subspace to the query in $\ell^1$ distance.

Object Recognition

Cannot find the paper you are looking for? You can Submit a new open access paper.