摘要:
阅读全文
摘要:
视角 阅读全文
摘要:
阅读全文
摘要:
阅读全文
摘要:
阅读全文
摘要:
http://www.gatsby.ucl.ac.uk/~gretton/coursefiles/RKHS2013_slides1.pdf RKHS: a function space with a very special structure 阅读全文
摘要:
https://en.wikipedia.org/wiki/Definite_quadratic_form https://www.math.utah.edu/~zwick/Classes/Fall2012_2270/Lectures/Lecture33_with_Examples.pdf 阅读全文
摘要:
http://www.gatsby.ucl.ac.uk/~gretton/coursefiles/lecture4_introToRKHS.pdf 阅读全文
摘要:
http://mathworld.wolfram.com/HilbertSpace.html A Hilbert space is a vector space with an inner product such that the norm defined by turns into a comp 阅读全文
摘要:
http://mp.weixin.qq.com/s/hImjVRMHHEC3SAepAoL2Tw 阅读全文
摘要:
Reproducing kernel Hilbert space Mapping the points to a higher dimensional feature space http://www.gatsby.ucl.ac.uk/~gretton/coursefiles/lecture4_in 阅读全文
摘要:
core action 阅读全文
摘要:
q 阅读全文
摘要:
airlock 阅读全文
摘要:
https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/funcZone_TKDE_Zheng.pdf Specifically, we introduce the concept of Latent Activity 阅读全文
摘要:
http://mp.weixin.qq.com/s/2NoYnwE_FxfhU9qCwH-nnA 用实际轮廓确定商圈的范围 阅读全文
摘要:
https://en.wikipedia.org/wiki/Frobenius_inner_product Frobenius norm 阅读全文
摘要:
In many cases, the smallness of the residual means that the approximation is close to the solution, i.e., 残差:结果的差值 误差:近似解与解的差值 https://en.wikipedia.or 阅读全文
摘要:
http://djt.qq.com/article/view/16 其主要思想就是把影响集中到一个点,然后再发散到一个面,出现意外情况后很容易就回退。 Gmail Labs是一个新特性橱窗,用户可以自己选择一些未正式发布的新特性进行体验,不喜欢可以关闭,在这个过程中,吃了螃蟹,也当了Google的小 阅读全文
摘要:
http://www.mit.edu/~9.520/scribe-notes/cl7.pdf https://en.wikipedia.org/wiki/Bayesian_interpretation_of_kernel_regularization the degree to which inst 阅读全文
摘要:
适定性问题 zh.wikipedia.org/wiki/適定性問題 数学术语适定性问题来自于哈达玛所给出的定义。他认为物理现象中的数学模型应该具备下述性质: 适定性问题的原型范例包括对于拉普拉斯方程的狄利克雷问题,以及给定初始条件的热传导方程式。在物理过程中解决的这些问题,也许被视为“自然”问题。相 阅读全文
摘要:
zh.wikipedia.org/wiki/數值微分 数值微分是数值方法中的名词,是用函数的值及其他已知资讯来估计一函数导数的算法。 http://mathworld.wolfram.com/NumericalDifferentiation.html Numerical differentiatio 阅读全文
摘要:
http://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html 阅读全文
摘要:
http://ufldl.stanford.edu/tutorial/supervised/OptimizationStochasticGradientDescent/ 阅读全文
摘要:
https://stats.stackexchange.com/questions/164876/tradeoff-batch-size-vs-number-of-iterations-to-train-a-neural-network It has been observed in practic 阅读全文
摘要:
https://ipfs.io/ipfs/QmXoypizjW3WknFiJnKLwHCnL72vedxjQkDDP1mXWo6uco/wiki/Convex_function.html Every norm is a convex function, by the triangle inequal 阅读全文
摘要:
http://mathworld.wolfram.com/Moore-PenroseMatrixInverse.html 显然,埃尔米特矩阵主对角线上的元素都是实数的,其特征值也是实数。对于只包含实数元素的矩阵(实矩阵),如果它是对称阵,即所有元素关于主对角线对称,那么它也是埃尔米特矩阵。也就是说, 阅读全文
摘要:
https://rdipietro.github.io/friendly-intro-to-cross-entropy-loss/ 【将输入转化为输出:概率分布】 When we develop a model for probabilistic classification, we aim to 阅读全文
摘要:
https://en.wikipedia.org/wiki/Claude_Shannon In 1948, the promised memorandum appeared as "A Mathematical Theory of Communication," an article in two 阅读全文
摘要:
http://cs231n.github.io/linear-classify/ 阅读全文
摘要:
every row of W is a classifier for one of the classes As we saw above, every row of W is a classifier for one of the classes. The geometric interpreta 阅读全文
摘要:
The hinge loss is a convex function, so many of the usual convex optimizers used in machine learning can work with it. It is not differentiable, but h 阅读全文
摘要:
http://www1.inf.tu-dresden.de/~ds24/lehre/ml_ws_2013/ml_11_hinge.pdf Two extremes: • Big 𝐶 → the loss is more important → better recognition rate but 阅读全文
摘要:
图灵的基本思想 图灵的基本思想是用机器来模拟人们用纸笔进行数学运算的过程,他把这样的过程看作下列两种简单的动作: 在纸上写上或擦除某个符号; 把注意力从纸的一处移动到另一处; 而在每个阶段,人要决定下一步的动作,依赖于(a)此人当前所关注的纸上某个位置的符号和(b)此人当前思维的状态。 在可计算性理 阅读全文
摘要:
crf 阅读全文
摘要:
http://mathworld.wolfram.com/FrobeniusNorm.html 阅读全文
摘要:
https://en.wikipedia.org/wiki/Norm_(mathematics) http://cs231n.github.io/classification/ 阅读全文