English
全部
搜索
图片
视频
地图
资讯
更多
购物
航班
旅游
笔记本
Top stories
Sports
U.S.
Local
World
Science
Technology
Entertainment
Business
More
Politics
过去 24 小时
时间不限
过去 1 小时
过去 7 天
过去 30 天
最佳匹配
最新
17 小时
HOLO 微云全息基于 Masked 预训练 Transformer 的红外光谱反卷积算法
在 Transformer 架构的基础上,微云全息基于“Masked 预训练”策略。这种策略最初源于 BERT 模型在语言理解任务中的成功经验,被证明能够有效捕捉序列中元素间的深层次关系。微云全息研究团队将其迁移到红外光谱数据建模中,提出了一种自监督学习框架,用于从大规模无标签的红外光谱数据中自动学习鲁棒特征。
一些您可能无法访问的结果已被隐去。
显示无法访问的结果
今日热点
Wins first national title
Italian fashion designer dies
Cause of death revealed
Donates $10M to Morris
Goes all-cash in WBD bid
Indiana judge, wife shot
Police hunt suspect
Trump slams UK deal
Judge allows new DHS policy
To launch blockchain platform
Australia closes beaches
2 dead in NY ice accident
Dolphins hire new head coach
Dismisses vice premier
Launches LA Senate bid
ISR bulldozes UNRWA HQ
120 ISIS members escape
Guthrie reveals 'new voice’
Emergency session on Iran
Earthquake rattles SoCal
Strengthen defense ties
Sworn in as NJ governor
Addresses UK Parliament
China’s population falls
RU drones, missiles hit Kyiv
Embassy in London OK’d
Injured in car accident
Threatens French wine tariff
Yellowfin tuna recalled
Agree to $1M, 1-year deal
Join striking nurses
Titans agree to hire Saleh?
Massive Michigan pileup
反馈