<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Jay_Tang的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Jay_Tang</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Jay_Tang]]></copyright><item><title><![CDATA[Decoupling Representation and Classifier for Long-Tailed Recognition 图像领域长尾分布分类问题方法]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/112343511</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/112343511</guid><author>Jay_Tang</author><pubDate>Fri, 08 Jan 2021 11:48:24 +0800</pubDate><description><![CDATA[文章目录IntroductionRecent DirectionsSampling StrategiesMethods of Learning ClassifiersClassifier Re-training (cRT)Nearest Class Mean classifier (NCM)τ\tauτ-normalized classifier (τ(\tau(τ-normalized)ExperimentsDatasetsEvaluation ProtocolResultsSampling matter]]></description><category></category></item><item><title><![CDATA[Relation Extraction 关系抽取综述]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/112130624</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/112130624</guid><author>Jay_Tang</author><pubDate>Sun, 03 Jan 2021 12:32:42 +0800</pubDate><description><![CDATA[文章目录往期文章链接目录Information Extraction v.s. Relation ExtractionExisting Works of REPattern-based MethodsStatistical Relation Extraction ModelsNeural Relation Extraction MethodsFuture DirectionsUtilizing More DataMethods to Denoise DS DataOpen Problem for Utili]]></description><category></category></item><item><title><![CDATA[Connecting the Dots: Document-level Neural Relation Extraction with Edge-oriented Graphs 关系抽取论文总结]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/111948895</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/111948895</guid><author>Jay_Tang</author><pubDate>Thu, 31 Dec 2020 08:47:50 +0800</pubDate><description><![CDATA[文章目录往期文章目录链接Relation Extraction (RE)document-level REIntuitionContributionOverview of Proposed ModelProposed ModelSentence Encoding LayerGraph construction LayerNode ConstructionEdge ConstructionInference LayerFirst StepSecond StepClassification LayerResul]]></description><category></category></item><item><title><![CDATA[跨语言学习归纳总结 Cross-Lingual Learning paper summary]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/111782773</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/111782773</guid><author>Jay_Tang</author><pubDate>Mon, 28 Dec 2020 14:14:15 +0800</pubDate><description><![CDATA[往期文章链接目录
文章目录往期文章链接目录Cross-lingual learningCross-lingual resourcesMultilingual distributional representationsEvaluation of multilingual distributional representationsParallel corpusWord AlignmentsMachine TranslationUniversal features (out of fashion)Biling]]></description><category></category></item><item><title><![CDATA[BERT and RoBERTa 知识点整理]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/108662688</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/108662688</guid><author>Jay_Tang</author><pubDate>Fri, 18 Sep 2020 12:11:09 +0800</pubDate><description><![CDATA[往期文章链接目录
文章目录往期文章链接目录BERT RecapOverviewBERT SpecificsThere are two steps to the BERT framework: pre-training and fine-tuningInput Output RepresentationsTasksresultsAblation studiesEffect of Pre-training TasksEffect of Model SizesReplication study of BERT p]]></description><category></category></item><item><title><![CDATA[What Does BERT Look At? An Analysis of BERT’s Attention 论文总结]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/108572586</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/108572586</guid><author>Jay_Tang</author><pubDate>Mon, 14 Sep 2020 09:38:27 +0800</pubDate><description><![CDATA[文章目录往期文章链接目录Before we startSurface-Level Patterns in AttentionProbing Individual Attention HeadsProbing Attention Head CombinationsClustering Attention Heads往期文章链接目录
往期文章链接目录
Before we start
In this post, I mainly focus on the conclusions the authors reach]]></description><category></category></item><item><title><![CDATA[The More You Know: Using Knowledge Graphs for Image Classification 论文总结]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/108353255</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/108353255</guid><author>Jay_Tang</author><pubDate>Wed, 02 Sep 2020 08:54:10 +0800</pubDate><description><![CDATA[文章目录往期文章链接目录OverviewIntuitionPrevious WorkMajor ContributionGraph Search Neural Network (GSNN)GSNN ExplanationThree networksDiagram visualizationAdvantageIncorporate the graph network into an image pipelineDatasetConclusion往期文章链接目录
往期文章链接目录
Overview
This p]]></description><category></category></item><item><title><![CDATA[Graph Convolutional Neural Network -  Spectral Convolution 图卷积神经网络 — 频域卷积详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/108166410</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/108166410</guid><author>Jay_Tang</author><pubDate>Mon, 24 Aug 2020 08:49:54 +0800</pubDate><description><![CDATA[Fourier Transform
Virtually everything in the world can be described via a waveform - a function of time, space or some other variable. For instance, sound waves, the price of a stock, etc. The Fourier Transform gives us a unique and powerful way of viewin]]></description><category></category></item><item><title><![CDATA[Graph Convolutional Neural Network - Spatial Convolution 图卷积神经网络 — 空域卷积详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/108115302</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/108115302</guid><author>Jay_Tang</author><pubDate>Thu, 20 Aug 2020 08:40:58 +0800</pubDate><description><![CDATA[Convolutional graph neural networks (ConvGNNs)
Convolutional graph neural networks (ConvGNNs) generalize the operation of convolution from grid data to graph data. The main idea is to generate a node vvv’s representation by
aggregating its own features xv\]]></description><category></category></item><item><title><![CDATA[Introduction to Graph Neural Network (GNN) 图神经网络入门详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/108047975</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/108047975</guid><author>Jay_Tang</author><pubDate>Mon, 17 Aug 2020 08:37:40 +0800</pubDate><description><![CDATA[文章目录往期文章链接目录NoteBackground and IntuitionIntro to Graph Neural NetworksGNNs FrameworkDefinitionRecurrent graph neural networks (RecGNNs)Introduction to RecGNNsBanach's Fixed Point TheoremRecGNNs v.s. RNNsLimitation of RecGNNsGated Graph Neural Networks (GGN]]></description><category></category></item><item><title><![CDATA[Kaggle: Jigsaw Multilingual Toxic Comment Classification Top Solutions 金牌思路总结]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/107926075</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/107926075</guid><author>Jay_Tang</author><pubDate>Tue, 11 Aug 2020 08:45:49 +0800</pubDate><description><![CDATA[Before we start
Two of my previous post might be helpful in getting a general understanding of the top solutions of this competition. Please feel free to check them out.

Knowledge Distillation clearly explained
Common Multilingual Language Modeling method]]></description><category></category></item><item><title><![CDATA[常见多语言模型详解 (M-Bert, LASER, MultiFiT, XLM)]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/107873888</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/107873888</guid><author>Jay_Tang</author><pubDate>Sat, 08 Aug 2020 07:45:48 +0800</pubDate><description><![CDATA[文章目录往期文章链接目录Ways of tokenizationWord-based tokenizationCharacter-based tokenizationSubword tokenizationExisting approaches for cross-lingual NLPOut-of-vocabulary (OOV) problem in mono/multi-lingual settingsM-BERT (Multi-lingual BERT)WHY MULTILINGUAL BERT W]]></description><category></category></item><item><title><![CDATA[Knowledge Distillation 知识蒸馏详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/107805661</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/107805661</guid><author>Jay_Tang</author><pubDate>Wed, 05 Aug 2020 06:59:54 +0800</pubDate><description><![CDATA[文章目录往期文章链接目录Shortcoming of normal neural networksGeneralization of InformationKnowledge DistillationA few DefinitionsGeneral idea of knowledge distillationTeacher and StudentTemperature &amp; EntropyTraining the Distil Model往期文章链接目录
往期文章链接目录
Currently, esp]]></description><category></category></item><item><title><![CDATA[Kaggle: Tweet Sentiment Extraction 方法总结 Part 2/2: 金牌思路总结]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/107059971</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/107059971</guid><author>Jay_Tang</author><pubDate>Wed, 01 Jul 2020 12:07:22 +0800</pubDate><description><![CDATA[Before we start
I attended two NLP competition in June, Tweet Sentiment Extraction and Jigsaw Multilingual Toxic Comment Classification, and I’m happy to be a Kaggle Expert from now on ????

Tweet Sentiment Extraction
Goal:
The objective in this competitio]]></description><category></category></item><item><title><![CDATA[Kaggle: Tweet Sentiment Extraction 方法总结 Part 1/2: 常用方法总结]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/107060211</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/107060211</guid><author>Jay_Tang</author><pubDate>Wed, 01 Jul 2020 12:06:29 +0800</pubDate><description><![CDATA[文章目录往期文章目录链接NoteBefore we startTweet Sentiment ExtractionWhat is the MAGIC?Common MethodsLabel SmoothingImplementation of Label SmoothingIn tensorflowIn pytorchMulti-sample dropoutImplementationStochastic Weight Averaging (SWA)Different learning rate setti]]></description><category></category></item><item><title><![CDATA[RNN, LSTM 图文详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/106542925</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/106542925</guid><author>Jay_Tang</author><pubDate>Thu, 04 Jun 2020 11:44:02 +0800</pubDate><description><![CDATA[文章目录往期文章链接目录Sequence DataWhy not use a standard neural network for sequence tasksRNNDifferent types of RNNsLoss function of RNNBackpropagation through timeVanishing gradients with RNNsAdvantages and Drawbacks of RNNLSTMTypes of gatesformulas and illustrati]]></description><category></category></item><item><title><![CDATA[Intro to Deep Learning & Backpropagation 深度学习模型介绍及反向传播算法推导详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/106346648</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/106346648</guid><author>Jay_Tang</author><pubDate>Tue, 26 May 2020 04:29:34 +0800</pubDate><description><![CDATA[文章目录Deep Neural Network往期文章链接目录Forward PropagationLoss functions of neural networkBack-propagationcompute ∂ℓ∂f(x)\frac{\partial \ell}{\partial f(x)}∂f(x)∂ℓ​compute ∂ℓ∂a(L+1)(x)\frac{\partial \ell}{\partial a^{(L+1)}(x)}∂a(L+1)(x)∂ℓ​compute  ∂ℓ∂h(k)(x)\frac]]></description><category></category></item><item><title><![CDATA[Log-Linear Model & CRF 条件随机场详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/106212417</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/106212417</guid><author>Jay_Tang</author><pubDate>Tue, 19 May 2020 13:15:11 +0800</pubDate><description><![CDATA[文章目录往期文章链接目录Log-Linear modelConditional Random Fields (CRF)Formal definition of CRFLog-linear model to linear-CRFInference problem for CRFLearning problem for CRFLearning problem for general Log-Linear modelLearning problem for CRFCompute Z(xˉ,w)Z(\bar x, ]]></description><category></category></item><item><title><![CDATA[GMM & K-means 高斯混合模型和K-means聚类详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/106153760</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/106153760</guid><author>Jay_Tang</author><pubDate>Sat, 16 May 2020 08:18:16 +0800</pubDate><description><![CDATA[往期文章链接目录
文章目录往期文章链接目录Gaussian mixture model (GMM)Interpretation from geometryInterpretation from mixture modelGMM Derivationset upSolve by MLESolve by EM AlgorithmK-means往期文章链接目录
Gaussian mixture model (GMM)
A Gaussian mixture model is a probabilistic mode]]></description><category></category></item><item><title><![CDATA[Probabilistic Graphical Model (PGM) 概率图模型框架详解]]></title><link>https://blog.youkuaiyun.com/Jay_Tang/article/details/106046038</link><guid>https://blog.youkuaiyun.com/Jay_Tang/article/details/106046038</guid><author>Jay_Tang</author><pubDate>Mon, 11 May 2020 02:50:37 +0800</pubDate><description><![CDATA[往期文章链接目录
Probabilistic Graphical Model (PGM)
Definition: A probabilistic graphical model is a probabilistic model for which a graph expresses the conditional dependence structure between random variables.
In general, PGM obeys following rules:
Sum&nbsp;Rul]]></description><category></category></item></channel></rss>