12 jiongjiongai

尚未进行身份认证

暂无相关简介

等级
TA的排名 1w+

gRPC

Protocolbufferdataisstructuredasmessages,whereeachmessageisasmalllogicalrecordofinformationcontainingaseriesofname-valuepairscalledfields.Servicemethod:UnaryRPCServers...

2018-08-21 17:13:03

Deep Learning Notes: Chapter 1 Introduction

前言最近开始读《DeepLearning》一书。这让我有了一个边读书边写笔记的动机:很有必要有一个笔记,能够让人很轻松流畅的读懂这本书的核心内容,至少可以把握住这本书的脉络。由于终究是英文表达更地道,因此该笔记都是节选自书中的原文。各位读者如果有建议或意见,欢迎留言。谢谢!DeepLearningChapter1IntroductionConceptDes...

2018-08-18 20:16:48

Pro Git Notes

GitisaDistributedVersionControlSystems(DVCSs).Clientsfullymirroroftherepository,includingitsfullhistory.

2018-08-12 01:38:23

多元函数的牛顿迭代法

f(X)=f(X0)+f′(X0)ΔX+12(ΔX)Tf″(X0)ΔXf(X)=f(X0)+f′(X0)ΔX+12(ΔX)Tf″(X0)ΔXf(X)=f(X_0)+f'(X_0)\DeltaX+\dfrac{1}{2}\left(\DeltaX\right)^Tf''(X_0)\DeltaX于是f′(X)=f′(X0)+f″(X0)ΔXf′(X)=f′...

2018-08-09 17:55:05

牛顿迭代法

若x=F(x)x=F(x)x=F(x)等价于f(x)=0f(x)=0f(x)=0则F(x)F(x)F(x)称为迭代函数。f(x)f(x)f(x)有二阶连续导数,且f′(x)≠0f′(x)≠0f'(x)\neq0则∀x0,x∈R,∀x0,x∈R,\forallx_0,x\in\mathbbR,若f(x0)=0f(x0)=0f(x_0)=0则...

2018-08-09 09:02:05

Nesterov Momentum

x_ahead=x+mu*v#evaluatedx_ahead(thegradientatx_aheadinsteadofatx)v=mu*v-learning_rate*dx_aheadx+=v=>x_prev=xv_prev=vx_ahead=x_prev+mu*v_prevv=mu*v_...

2018-08-09 08:19:11

CS231n Note

CS231nNoteConceptsConceptDescriptionImageClassificationObjectDetectionActionClassificationImageCaptioningSemanticSegmentationPerceptual...

2018-08-04 20:56:43

Clockwise/Spiral Rule to parse C declaration

http://c-faq.com/decl/spiral.anderson.html

2018-05-01 18:32:17

推荐系统

推荐方式社会化推荐(socialrecommendation)基于内容的推荐(content-basedfiltering)协同过滤(collaborativefiltering)推荐系统评测推荐系统试验方法离线试验用户调查在线试验(AB测试)评测指标用户满意度预测准确度覆盖率多样性新颖性惊喜度(serendipity)信任度实时性...

2018-04-25 21:52:44

机器学习的求导公式

机器学习的求导公式损失函数的求导公式设loss(X)loss⁡(X)\operatorname{loss}\left(X\right)为单个样本XXX的损失函数,A=g(Z)=⎛⎝⎜⎜g(z1)⋮g(zn)⎞⎠⎟⎟A=g(Z)=(g⁡(z1)⋮g⁡(zn))A=g\left(Z\right)=\begin{pmatrix}\operatorname{g...

2018-04-18 12:30:10

Recurrent Neural Networks

ExamplesofSequenceDataSpeechRecognitionMusicGenerationSentimentClassificationDNASequenceAnalysisMachineTranslationVideoActivityRecognitionNameEntityRecognitionNotation...

2018-04-16 06:28:37

Neural Style Transfer

ConceptContentC+StyleS=GeneratedimageGWhatareDeepConvNetLearning?Moreabstractfeaturesindeeperlayer.CostFunctionloss(G;C,S)=αlosscontent(S,G)+βlossstyle(C,G)loss⁡(G;C,...

2018-04-16 00:04:20

Face Recognition

FaceVerificationvsFaceRecognitionNameInputOutputDescriptionFaceVerificationImageandName/IDIstheimagethepersonwiththisgivenID?FaceRecognitionIma...

2018-04-15 19:39:30

Object Detection

ConceptsNameDescriptionyyyObjectClassificationAtmostoneobjecty=⎛⎝⎜c1c2c3⎞⎠⎟y=(c1c2c3)y=\begin{pmatrix}c_1\\c_2\\c_3\end{pmatrix}ObjectLocalizationAtmost...

2018-04-15 19:17:33

Convolutional Neural Networks

PaddingOutputDimensionn+2p−f+1n+2p−f+1n+2p-f+1PaddingTypesValid:p=0p=0p=0Same:n+2p−f+1=n⇒p=f−12n+2p−f+1=n⇒p=f−12n+2p-f+1=n\Rightarrowp=\dfrac{f-1}{2}Str...

2018-04-13 01:34:11

Learning from Multiple tasks

WhereTransferLearningfromAtoBMakesSenseTaskAandBhavethesameinputX.YouhavealotmoredataforAthanB.LowlevelfeaturesinAcouldbehelpfulforlearningB.WhereMulti-task...

2018-04-12 23:37:42

Bias and Variance with Mismatched Distributions

BiasandVariancewithMismatchedDistributions

2018-04-12 22:08:00

Softmax Function

SigmoidFunctionsigmoid(z)=11−e−zsigmoid⁡(z)=11−e−z\operatorname{sigmoid}(z)=\dfrac{1}{1-e^{-z}}SoftmaxFunctionsoftmax(zi;Z)=ezi∑i=1nezi,1≤i≤nsoftmax⁡(zi;Z)=ezi∑i=1nezi,1≤i≤n\operato...

2018-04-11 22:13:26

Momentum, RMSprob and Adam

GradientDescentwithMomentumComputeexponentiallyweighedaverageofgradient,andusethegradienttoupdateweights.AlgorithmOniterationt:ComputedWd⁡W\operatorname{d}Wanddbd⁡b\op...

2018-04-11 02:03:54

Exponentially Weighted Averages

ExponentiallyWeightedAveragesvt=βvt−1+(1−β)θtvt=βvt−1+(1−β)θtv_{t}=\betav_{t-1}+\left(1-\beta\right)\theta_{t}=β[βvt−2+(1−β)θt−1]+(1−β)θt=β[βvt−2+(1−β)θt−1]+(1−β)θt=\beta\lef...

2018-04-11 00:09:48

查看更多

勋章 我的勋章
    暂无奖章