开课单位--魏茨曼科学研究所
11
12
13
14
15
16
17
18
19
20
![](functions/showpic.php?filename=2016113002585198.png)
Using linking features in learning non-parametric part models[在学习非参数零件模型中的链接特征]
Stefan Carlsson;Antonio Torralba;Leonid Karlinsky(魏茨曼科学研究所) We present an approach to the detection of parts of highly deformable objects, such as the human body. Instead of using kinematic constraints on relat...
热度:24
Stefan Carlsson;Antonio Torralba;Leonid Karlinsky(魏茨曼科学研究所) We present an approach to the detection of parts of highly deformable objects, such as the human body. Instead of using kinematic constraints on relat...
热度:24
![](functions/showpic.php?filename=2019072604302458.png)
Efficient Online Learning via Randomized Rounding[通过随机舍入实现高效的在线学习]
Ohad Shamir(魏茨曼科学研究所) Most online algorithms used in machine learning today are based on variants of mirror descent or follow-the-leader. In this paper, we present an onlin...
热度:91
Ohad Shamir(魏茨曼科学研究所) Most online algorithms used in machine learning today are based on variants of mirror descent or follow-the-leader. In this paper, we present an onlin...
热度:91
![](functions/showpic.php?filename=2019071008510650.png)
Diffusion Maps, Spectral Clustering and Reaction Coordinates of Dynamical Systems[动力系统的扩散映射,谱聚类和反应坐标]
Boaz Nadler(魏茨曼科学研究所) A central problem in data analysis is the low dimensional representation of high dimensional data, and the concise description of its underlying geome...
热度:129
Boaz Nadler(魏茨曼科学研究所) A central problem in data analysis is the low dimensional representation of high dimensional data, and the concise description of its underlying geome...
热度:129
![](functions/showpic.php?filename=2019063008181619.jpg)
A Universal Kernel for Learning Regular Languages[学习常规语言的通用核心]
Leonid Kontorovich(魏茨曼科学研究所) We give a universal kernel that renders all the regular languages linearly separable. We are not able to compute this kernel efficiently and conject...
热度:18
Leonid Kontorovich(魏茨曼科学研究所) We give a universal kernel that renders all the regular languages linearly separable. We are not able to compute this kernel efficiently and conject...
热度:18
![](functions/showpic.php?filename=2019030707563314.png)
Collaborative Filtering with the Trace Norm: Learning, Bounding, and Transducing [跟踪规范的协作过滤:学习、边界和转换]
Ohad Shamir(魏茨曼科学研究所) Trace-norm regularization is a widely-used and successful approach for collaborative filtering and matrix completion. However, its theoretical unders...
热度:86
Ohad Shamir(魏茨曼科学研究所) Trace-norm regularization is a widely-used and successful approach for collaborative filtering and matrix completion. However, its theoretical unders...
热度:86
![](functions/showpic.php?filename=2019042508450328.png)
A metric notion of dimension and its applications to learning[维度的度量概念及其在学习中的应用]
Robert Krauthgamer(魏茨曼科学研究所) Let us define the dimension of a metric space as the minimum k>0 such that every ball in the metric space can be covered by 2^k balls of half the r...
热度:84
Robert Krauthgamer(魏茨曼科学研究所) Let us define the dimension of a metric space as the minimum k>0 such that every ball in the metric space can be covered by 2^k balls of half the r...
热度:84
![](functions/showpic.php?filename=2019042108354264.png)
Learning to Classify with Missing and Corrupted Features[学习使用缺失和损坏的功能进行分类]
Ohad Shamir(魏茨曼科学研究所) After a classifier is trained using a machine learning algorithm and put to use in a real world system, it often faces noise which did not appear in t...
热度:60
Ohad Shamir(魏茨曼科学研究所) After a classifier is trained using a machine learning algorithm and put to use in a real world system, it often faces noise which did not appear in t...
热度:60
![](functions/showpic.php?filename=2019041411201197.png)
From Bandits to Experts : On the Value of More Information[从土匪到专家:更多信息的价值]
Ohad Shamir(魏茨曼科学研究所) Learning from Experts and Multi-armed Bandits are two of the most common settings studied in online learning. Whereas the first setting assumes that t...
热度:32
Ohad Shamir(魏茨曼科学研究所) Learning from Experts and Multi-armed Bandits are two of the most common settings studied in online learning. Whereas the first setting assumes that t...
热度:32
![](functions/showpic.php?filename=2019032008171247.png)
“Clustering by Composition” for Unsupervised Discovery of Image Categories[“按构图聚类”用于无监督发现图像类别]
Tinne Tuytelaars, Alon Faktor, Serge J. Belongie(魏茨曼科学研究所) We define a good image cluster as one in which images can be easily composed (like a puzzle) using pieces from each other, while are difficult to comp...
热度:51
Tinne Tuytelaars, Alon Faktor, Serge J. Belongie(魏茨曼科学研究所) We define a good image cluster as one in which images can be easily composed (like a puzzle) using pieces from each other, while are difficult to comp...
热度:51
![](functions/showpic.php?filename=2019031209063816.png)
The chains model for detecting parts by their context[基于上下文的零件检测链模型]
Leonid Karlinsky(魏茨曼科学研究所) Detecting an object part relies on two sources of information - the appearance of the part itself, and the context supplied by surrounding parts. In t...
热度:60
Leonid Karlinsky(魏茨曼科学研究所) Detecting an object part relies on two sources of information - the appearance of the part itself, and the context supplied by surrounding parts. In t...
热度:60