RI: Small: A Study of New Aggregate Losses for Machine Learning

RI：小：机器学习新总损失的研究

基本信息

批准号：
2008532
负责人：
Siwei Lyu
金额：
$ 45万
依托单位：
SUNY at Albany
依托单位国家：
美国
项目类别：
Standard Grant
财政年份：
2020
资助国家：
美国
起止时间：
2020-10-01 至 2020-12-31
项目状态：
已结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=2008532&HistoricalAwards=false
关键词：
RI Small Study New Aggregate

项目摘要

Machine learning is instrumental for the recent advances in AI and big data analysis. They have been used in almost every area of computer science and many fields of natural sciences, engineering, and social sciences. The central task of machine learning is to “train” a model, which entails seeking models that minimize certain performance metrics over a set of training examples. Such performance metrics are termed as the aggregate losses, which are to be distinguished from the individual losses that measures the quality of the model on a single training example. As the link between the training data and the model to be learned, the aggregate loss is a fundamental component in machine learning algorithms, and its theoretical and practical significance warrants a comprehensive and systematic study. The proposed work will focus on several fundamental research questions concerning the aggregate loss: are there any other types of aggregate loss beyond the average individual losses?; if so, what will be a general abstract formulation of these new aggregate loss?; how can the new aggregate losses be adapted to different machine learning problems?; and what are the statistical and computational behaviors of machine learning algorithms using the general aggregate losses?. The technical aims of the project are divided into four interrelated thrusts. The first thrust explores new types of rank-based aggregate losses for binary classification and study efficient algorithms optimizing learning objectives formed based upon them. The new aggregate losses will be applied to problems such as object detection, where rank-based evaluation metric is used dominantly. The second thrust aims to deepen our understanding of the binary classification algorithms developed using the rank-based aggregate losses and will be focused on a study of their statistical theories such as generalization and consistency. The third thrust will extend the study of new types of aggregate losses to other supervised problems (multi-class and multi-label learning and supervised metric learning) and unsupervised learning. The fourth thrust dedicates to the theoretical aspects of aggregate losses, in which an aggregate loss will be abstracted as a set function that maps the ensemble of individual losses to a number. This abstraction will be exploited to study the properties of new aggregate losses that make them superior than the average loss and propose new aggregate losses beyond rank-based ones.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

机器学习对自然科学，工程学和社会科学领域的最新进展和机器学习的核心是“训练”模型，这需要训练示例实际的意义需要一项全面的研究，以超出平均个人损失的其他类型的损失？使用该项目的技术目标的机器学习算法的统计和计算是四个相互关联的推力第二个推力是使用基于等级的损失开发的HMS，并将重点放在对统计线程的研究中，作为概括和一致性。学习。第四个推力专用于总损失的理论方面，将抽象为将个体损失的合奏映射到一个数字上。并提出了超出基于排名的新的损失。这rd Eflects使用Toundation Ts Review标准，NSF的NSF的suthy suthy suthy suthy suthy。

项目成果

期刊论文数量（17）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Learning by Minimizing the Sum of Ranked Range

DOI：
发表时间：
2020-10
期刊：
ArXiv
影响因子：
0
作者：
Shu Hu;Yiming Ying;Xin Wang;Siwei Lyu
通讯作者：
Shu Hu;Yiming Ying;Xin Wang;Siwei Lyu

Stability and differential privacy of stochastic gradient descent for pairwise learning with non-smooth loss

非平滑损失成对学习的随机梯度下降的稳定性和差分隐私

DOI：
发表时间：
2021
期刊：
Proceedings of the 24th International Conference on Artificial Intelligence and Statistics (AISTATS
影响因子：
0
作者：
Yang, Zhenhuan;Lei, Yunwen;Lyu, Siwei;Ying, Yiming
通讯作者：
Ying, Yiming

Unmixing Biological Fluorescence Image Data with Sparse and Low-Rank Poisson Regression

DOI：
10.1101/2023.01.06.523044
发表时间：
2023-01
期刊：
bioRxiv
影响因子：
0
作者：
Ruogu Wang;A. Lemus;Colin M. Henneberry;Yiming Ying;Yunlong Feng;A. Valm
通讯作者：
Ruogu Wang;A. Lemus;Colin M. Henneberry;Yiming Ying;Yunlong Feng;A. Valm

Differentially Private SGDA for Minimax Problems

DOI：
发表时间：
2022-01
期刊：
ArXiv
影响因子：
0
作者：
Zhenhuan Yang;Shu Hu;Yunwen Lei;Kush R. Varshney;Siwei Lyu;Yiming Ying
通讯作者：
Zhenhuan Yang;Shu Hu;Yunwen Lei;Kush R. Varshney;Siwei Lyu;Yiming Ying

Stability and Generalization for Markov Chain Stochastic Gradient Methods

DOI：
10.48550/arxiv.2209.08005
发表时间：
2022-09
期刊：
ArXiv
影响因子：
0
作者：
Puyu Wang;Yunwen Lei;Yiming Ying;Ding-Xuan Zhou
通讯作者：
Puyu Wang;Yunwen Lei;Yiming Ying;Ding-Xuan Zhou

DOI：
{{ item.doi }}
发表时间：
{{ item.publish_year }}
期刊：
{{ item.journal_name }}
影响因子：
{{ item.factor }}
作者：
{{ item.authors }}
通讯作者：
{{ item.author }}

数据更新时间：{{ journalArticles.updateTime }}

作者：
{{ item.author }}

数据更新时间：{{ monograph.updateTime }}

作者：
{{ item.author }}

数据更新时间：{{ sciAawards.updateTime }}

作者：
{{ item.author }}

数据更新时间：{{ conferencePapers.updateTime }}

作者：
{{ item.author }}

数据更新时间：{{ patent.updateTime }}

Siwei Lyu其他文献

Deep Constrained Low-Rank Subspace Learning for Multi-View Semi-Supervised Classification

用于多视图半监督分类的深度约束低秩子空间学习

DOI：
10.1109/lsp.2019.2923857
发表时间：
2019
期刊：
IEEE Signal Processing Letters（二区）
影响因子：
0
作者：
Zhe Xue;Junping Du;Dawei Du;Guorong Li;Qingming Huang;Siwei Lyu
通讯作者：
Siwei Lyu

Countering JPEG anti-forensics based on noise level estimation

基于噪声水平估计的 JPEG 反取证对抗

DOI：
10.1007/s11432-016-0426-1
发表时间：
2017-08
期刊：
Science in China
影响因子：
0
作者：
Hui Zeng;Xiangui Kang;Jingjing Yu;Siwei Lyu
通讯作者：
Siwei Lyu

Online Deformable Object Tracking Based on Structure-Aware Hyper-Graph

基于结构感知超图的在线变形目标跟踪

DOI：
10.1109/tip.2016.2570556
发表时间：
2016-08
期刊：
IEEE Transactions on Image Processing
影响因子：
10.6
作者：
Dawei Du;Honggang Qi;Wenbo Li;Longyin Wen;Qingming Huang;Siwei Lyu
通讯作者：
Siwei Lyu

Vertebral artery course variation leading to an insufficient proximal anchoring area for thoracic endovascular aortic repair.

椎动脉走行变化导致胸主动脉腔内修复的近端锚固区域不足。

DOI：
10.1177/17085381221140319
发表时间：
2022
期刊：
Vascular
影响因子：
1.1
作者：
Zuanbiao Yu;Siwei Lyu;Dehai Lang;Di Wang;Songjie Hu;Xiaoliang Yin;Yunpeng Ding;Chunbo Xu;Chen Lin;Jiangnan Hu
通讯作者：
Jiangnan Hu

Nonnegative matrix factorization with matrix exponentiation

DOI：
10.1109/icassp.2010.5494975
发表时间：
2010-03
期刊：
2010 IEEE International Conference on Acoustics, Speech and Signal Processing
影响因子：
0
作者：
Siwei Lyu
通讯作者：
Siwei Lyu