papers

Publications (23)

cs.LG2023

Does generalization performance of regularization learning depend on ? A negative example

Shaobo Lin, Chen Xu, Jingshan Zeng +1

-regularization has been demonstrated to be an attractive technique in machine learning and statistical modeling. It attempts to improve the generalization (prediction) capabi…

cs.IT2013

Sparse Solution of Underdetermined Linear Equations via Adaptively Iterative Thresholding

Jinshan Zeng, Shaobo Lin, Zongben Xu

Finding the sparset solution of an underdetermined system of linear equations has attracted considerable attention in recent years. Among a large number of algorithms, itera…

cs.CV2023

Explore the Power of Synthetic Data on Few-shot Object Detection

Shaobo Lin, Kun Wang, Xingyu Zeng +1

Few-shot object detection (FSOD) aims to expand an object detector for novel categories given only a few instances for training. The few training samples restrict the performance o…

cs.CV2023

An Effective Crop-Paste Pipeline for Few-shot Object Detection

Shaobo Lin, Kun Wang, Xingyu Zeng +1

Few-shot object detection (FSOD) aims to expand an object detector for novel categories given only a few instances for training. However, detecting novel categories with only a few…

cs.LG2023

Model selection of polynomial kernel regression

Shaobo Lin, Xingping Sun, Zongben Xu +1

Polynomial kernel regression is one of the standard and state-of-the-art learning strategies. However, as is well known, the choices of the degree of polynomial kernel and the regu…

cs.CV2022

A Unified Framework with Meta-dropout for Few-shot Learning

Shaobo Lin, Xingyu Zeng, Rui Zhao

Conventional training of deep neural networks usually requires a substantial amount of data with expensive human annotations. In this paper, we utilize the idea of meta-learning to…

math.OC2015

Sparse Regularization: Convergence Of Iterative Jumping Thresholding Algorithm

Jinshan Zeng, Shaobo Lin, Zongben Xu

In recent studies on sparse modeling, non-convex penalties have received considerable attentions due to their superiorities on sparsity-inducing over the convex counterparts. Compa…

cs.LG2014

Learning and approximation capability of orthogonal super greedy algorithm

Jian Fang, Shaobo Lin, Zongben Xu

We consider the approximation capability of orthogonal super greedy algorithms (OSGA) and its applications in supervised learning. OSGA is concerned with selecting more than one at…

cs.LG2016

Constructive neural network learning

Shaobo Lin, Jinshan Zeng, Xiaoqin Zhang

In this paper, we aim at developing scalable neural network-type learning systems. Motivated by the idea of "constructive neural networks" in approximation theory, we focus on "con…

cs.LG2014

Greedy metrics in orthogonal greedy learning

Lin Xu, Shaobo Lin, Jinshan Zeng +1

Orthogonal greedy learning (OGL) is a stepwise learning scheme that adds a new atom from a dictionary via the steepest gradient descent and build the estimator via orthogonal proje…

cs.LG2015

Shrinkage degree in -re-scale boosting for regression

Lin Xu, Shaobo Lin, Yao Wang +1

Re-scale boosting (RBoosting) is a variant of boosting which can essentially improve the generalization performance of boosting learning. The key feature of RBoosting lies in intro…

cs.LG2014

Is Extreme Learning Machine Feasible? A Theoretical Assessment (Part II)

Shaobo Lin, Xia Liu, Jian Fang +1

An extreme learning machine (ELM) can be regarded as a two stage feed-forward neural network (FNN) learning system which randomly assigns the connections with and within hidden neu…

cs.LG2014

Learning rates of coefficient regularization learning with Gaussian kernel

Shaobo Lin, Jinshan Zeng, Jian Fang +1

Regularization is a well recognized powerful strategy to improve the performance of a learning machine and regularization schemes with are central in use. It is…

cs.LG2018

Learning through deterministic assignment of hidden parameters

Jian Fang, Shaobo Lin, Zongben Xu

Supervised learning frequently boils down to determining hidden and bright parameters in a parameterized hypothesis space based on finite input-output samples. The hidden parameter…

cs.LG2016

Greedy Criterion in Orthogonal Greedy Learning

Lin Xu, Shaobo Lin, Jinshan Zeng +2

Orthogonal greedy learning (OGL) is a stepwise learning scheme that starts with selecting a new atom from a specified dictionary via the steepest gradient descent (SGD) and then bu…

cs.LG2016

Divide and Conquer Local Average Regression

Xiangyu Chang, Shaobo Lin, Yao Wang

The divide and conquer strategy, which breaks a massive data set into a se- ries of manageable data blocks, and then combines the independent results of data blocks to obtain a fin…

cs.LG2015

Re-scale boosting for regression and classification

Shaobo Lin, Yao Wang, Lin Xu

Boosting is a learning scheme that combines weak prediction rules to produce a strong composite estimator, with the underlying intuition that one can obtain accurate prediction rul…

math.OC2014

A Cyclic Coordinate Descent Algorithm for lq Regularization

Jinshan Zeng, Zhimin Peng, Shaobo Lin +1

In recent studies on sparse modeling, () regularization has received considerable attention due to its superiorities on sparsity-inducing and bias reduction over the $…

math.OC2019

Global Convergence of Block Coordinate Descent in Deep Learning

Jinshan Zeng, Tim Tsz-Kit Lau, Shaobo Lin +1

Deep learning has aroused extensive attention due to its great empirical success. The efficiency of the block coordinate descent (BCD) methods has been recently demonstrated in dee…

cs.CV2023

Explore the Power of Dropout on Few-shot Learning

Shaobo Lin, Xingyu Zeng, Rui Zhao

The generalization power of the pre-trained model is the key for few-shot deep learning. Dropout is a regularization technique used in traditional deep learning methods. In this pa…

math.NA2014

Regularization: Convergence of Iterative Half Thresholding Algorithm

Jinshan Zeng, Shaobo Lin, Yao Wang +1

In recent studies on sparse modeling, the nonconvex regularization approaches (particularly, regularization with ) have been demonstrated to possess capability o…

math.NA2015

A Gauss-Seidel Iterative Thresholding Algorithm for lq Regularized Least Squares Regression

Jinshan Zeng, Zhimin Peng, Shaobo Lin

In recent studies on sparse modeling, () regularized least squares regression (LS) has received considerable attention due to its superiorities on sparsity-induci…

cs.LG2015

Nonparametric regression using needlet kernels for spherical data

Shaobo Lin

Needlets have been recognized as state-of-the-art tools to tackle spherical data, due to their excellent localization properties in both spacial and frequency domains. This paper c…