momoka9的博客

颜锐, 梁智勇, 李锦涛, 任菲. 基于深度学习和H&E染色病理图像的肿瘤相关指标预测研究综述[J]. 计算机科学, 2022, 49(2): 69-82.YAN Rui, LIANG Zhi-yong, LI Jin-tao, REN Fei. Predicting Tumor-related Indicators Based on Deep Learning and H&E Stained Pathological Images:A Survey[J]. Computer Scie.

1、BoostMIS: Boosting Medical Image Semi-supervised Learning with Adaptive Pseudo Labeling and Informative Active Annotation（CVPR 2022）BoostMIS：利用自适应伪标签和信息性主动注释促进医学图像半监督学习在本文中，我们提出了一个新颖的半监督学习（SSL）框架，名为BoostMIS，它结合了自适应伪标签和信息性主动注释，以释放医学图像SSL模型的潜力。(1）Boost

Fine-Grained Visual Categorization via Multi-stage Metric Learning细粒度视觉分类（FGVC）是指将对象归入从属类而不是基本类。FGVC的一个主要挑战是两个问题的共同存在。1）许多从属类高度相关，难以区分；2）存在较大的类内变化（例如，由于物体的姿势）。本文提出通过距离度量学习（DML）明确地解决上述两个问题。DML通过学习嵌入来解决第一个问题，使来自同一类别的数据点被拉到一起，而来自不同类别的数据点应该被相互推开；它通过允许同一类别的.

Top-k Hierarchical ClassificationThirty-First AAAI Conference on Artificial IntelligenceDespite that the fact that both top-k classification and hierarchical classification have gained increasing interests, the two problems have always been studied s.

Many-Class Few-Shot Learning on Multi-Granularity Class HierarchyPublished in:IEEE Transactions on Knowledge and Data Engineering(Volume: 34,Issue: 5, May 1 2022)many-class few-shot? 类别很多，但是每个类别的样本数量很少（less then 100）1、在类别多且各个类别样本多的情况下，DNN具有非凡的力...

COARSE-TO-FINE CURRICULUM LEARNING FOR CLASSIFICATIONPublished as a workshop paper at “Bridging AI and Cognitive Science” (ICLR 2020)In this paper, we propose an algorithm for learning and using a curriculum in the context of machine learning没看懂...

Cascade Learning for Mobile Malware Families Detection through Quality and Android MetricsPublished in:2019 International Joint Conference on Neural Networks (IJCNN)一、摘要设计了一个级联学习器，其中级联的第一个分类器执行粗粒度分析（它区分恶意软件和合法应用程序），而第二个分类器执行细粒度分析（它旨在识别恶意软件家族）。在合法样..

1、Coarse-to-fine classification for diabetic retinopathy grading using convolutional neural network使用卷积神经网络对糖尿病视网膜病变分级进行从粗到细的分类。Artificial Intelligence in MedicineVolume 108, August 2020, 101936亮点首次提出了一个分层的从粗到细的糖尿病视网膜病变（DR）网络（CF-DRNet），该网络可以利用卷积神