scNym-用于单细胞分类的半监督对抗神经网络 scNym是一个神经网络模型,用于根据单细胞分析数据(例如scRNA-seq)预测细胞类型,并从这些模型中得出细胞类型表示形式。 尽管细胞类型分类是主要的用例,但是这些模型可以将单个细胞概况映射到任意输出类别(例如实验条件)。 我们已经在Genome Research的最新论文中详细描述了scNym 。 如果您发现此工具有用,请引用我们的工作。 我们也有一个研究网站,介绍scNym简报- 用于单细胞分类的半监督对抗神经网络。 雅各布·金梅尔(Jacob C.Kimmel)和大卫·凯利(David R.Kelley)。 基因组研究。 2021. doi: : BibTeX @article{kimmel_scnym_2021, title = {Semi-supervised adversarial neural networ
1
Semi-supervised Anomaly Detection using AutoEncoders PDF全文翻译,属于缺陷检测的文档,适合于研究目标检测方面的研究者
2022-04-30 09:05:02 355KB 文档资料
1
Semi-supervised Anomaly Detection using AutoEncoders全文英文注解,适合于英文较好的研究者,看英文的文档更有味道一些
2022-04-30 09:05:02 1.48MB 文档资料
1
SEMI E30-1103 GENERIC MODEL FOR COMMUNICATIONS AND CONTROL OF MANUFACTURING EQUIPMENT (GEM) SEMI E30.1-0200 INSPECTION AND REVIEW SPECIFIC EQUIPMENT MODEL (ISEM) SEMI E30.5-0302 SPECIFICATION FOR METROLOGY SPECIFIC EQUIPMENT MODEL
2022-04-26 17:03:10 11.61MB E30 E30.1 E30.5
1
SEMI E090標準規範,是半導體最重點的規範,全部半導體都畢需尊守此規範 Abstract This Standard was technically approved by the Information & Control Global Technical Committee. This edition was approved for publication by the global Audits and Reviews Subcommittee on August 17, 2018. Available at www.semiviews.org and www.semi.or
2022-04-25 18:28:50 1.04MB semi
1
SEMI E040-0705_PMS半導體標準規範 ,所有半導體公司都必需尊守此規範 About the SEMI Standards Program Why standards? The SEMI International Standards Program is one of the key services offered by Semiconductor Equipment and Materials International (SEMI) for the benefit of the worldwide semiconductor, photovoltaic (PV), LED, MEMS and flat panel display (FPD) industries. Standards offer a way to meet the challenges of increasing productivity while enabling business opportunities around the globe. The program, started over 40 years ago in North America, was expanded in 1985 to include programs in Europe and Japan, and now also has technical committees in China, Korea and Taiwan. In 1997, the Program was expanded to cover other areas with activity in these industries among suppliers and users. The program operates as a neutral forum for the exchange of information among suppliers and users resulting in the production of timely and technically accurate specifications and other standards of economic importance to the industry. It is a vehicle for networking, partnering, and professional growth. Over 5,000 technologists worldwide, representing both device manufacturers and equipment and materials suppliers, participate in the program. These individuals work toward resolving a variety of process and product related issues in both the front and back-end areas in semiconductor, photovoltaic, and flat panel display manufacturing.
2022-04-25 18:28:04 933KB semi
1
设备自动化协会标准文档,半导体制造行业自动运输系统需要遵守的通讯标准。自99年在E23的基础上制定了E84后,E84逐渐成为主流,开发必备。
2022-04-25 18:27:25 1.69MB SEMI E84
1
End-to-End Semi-Supervised Learning for Video Action Detection的阅读涂鸦 CVPR 2022 task:端到端的半监督视频动作检测方法
2022-04-06 03:11:27 10.66MB 论文阅读 深度学习
1
SEMI E5-1104 SEMI EQUIPMENT COMMUNICATIONS STANDARD 2 MESSAGE CONTENT (SECS-II)
2022-03-29 00:27:53 2.66MB SEMI E5 SECSII
1
协同训练是半监督的一个很好的范例,它要求用两个特征视图来描述数据集。 许多协同训练算法都有一个显着的特征:应以高置信度预测所选的未标记实例,因为高置信度得分通常表示相应的预测是正确的。 不幸的是,使用这些高置信度未标记实例并不总是能够提高分类性能。 本文提出了一种新的半监督学习算法,结合了联合训练和主动学习的优点。 该算法根据高置信度和最近邻两个准则应用协同训练来选择最可靠的实例,以提高分类器的性能,并利用具有人类注释能力的信息量最大的实例来提高分类性能。 在几个UCI数据集和自然语言处理任务上进行的实验表明,我们的方法在牺牲相同的人工量的情况下实现了更显着的改进。
2022-03-25 15:37:30 2.08MB Semi-supervised learning; Co-training; Confidence
1