=========== base model ===========

https://github.com/52Pig/paper_reading/blob/master/logistic-regression-ch12.pdf

https://github.com/52Pig/paper_reading/blob/master/fpgrowth.pdf

https://github.com/52Pig/paper_reading/blob/master/randomforest2001.pdf

https://github.com/52Pig/paper_reading/blob/master/GBDT-GreedyFuncApproxSS.pdf

https://github.com/52Pig/paper_reading/blob/master/xgboost_1603.02754.pdf

Rendle2010FM

https://github.com/52Pig/paper_reading/blob/master/Field-aware%20Factorization%20Machines%20for%20CTR%20Prediction.pdf

https://github.com/52Pig/paper_reading/blob/master/DeepFM-0239.pdf

https://github.com/52Pig/paper_reading/blob/master/SVDPaper.pdf

https://github.com/52Pig/paper_reading/blob/master/practical-lessons-from-predicting-clicks-on-ads-at-facebook.pdf

https://github.com/52Pig/paper_reading/blob/master/Deep%20Neural%20Networks%20for%20YouTube%20Recommendations.pdf

https://github.com/52Pig/paper_reading/blob/master/Deep%20Interest%20Network%20for%20Click-Through%20Rate%20Prediction.1706.06978.pdf

===============NLP=====================

https://github.com/52Pig/paper_reading/blob/master/language-models.pdf

https://github.com/52Pig/paper_reading/blob/master/textcnn_1408.5882.pdf

https://github.com/52Pig/paper_reading/blob/master/cikm2013_DSSM_fullversion.pdf

https://github.com/52Pig/paper_reading/blob/master/2017.%20CIKM%20HIN2Vec.pdf

https://github.com/52Pig/paper_reading/blob/master/node2vec%20Scalable%20Feature%20Learning%20for%20Networks.pdf

https://github.com/52Pig/paper_reading/blob/master/1409.3215.seq2seqwithneuralnetwork.pdf

https://github.com/52Pig/paper_reading/blob/master/bert_1810.04805.pdf

时间复杂度L*log(L); google开源，数据结构上变化，用到bucket桶和hash，self-attention计算的时候做优化。 Reformer：The Efficient Transformer。 https://github.com/52Pig/paper_reading/blob/master/REFORMER%20THE%20EFFICIENT%20TRANSFORMER.pdf

时间复杂度L; facebook ai开源，矩阵变换。矩阵操作包含x与w点积，w和qkv矩阵运算，是在x与w过程中进行拆解，x和L矩阵相乘，L和w矩阵相乘，这两部分进行优化，将复杂度转成线性的。Linformer：Self-Attention with Linear Complexity https://github.com/52Pig/paper_reading/blob/master/Linformer%20Self-Attention%20with%20Linear%20Complexity.pdf

https://github.com/52Pig/paper_reading/blob/master/Memory%20Transformer.2006.11527.pdf

bert针对推理速度的优化 https://github.com/52Pig/paper_reading/blob/master/2004.02178.pdf

################################## ACL 2020 优质论文 ################################## Don't Stop Pretraining: Adapt Language Models to Domains and Tasks. https://github.com/52Pig/paper_reading/blob/master/Beyond%20Accuracy%20Behavioral%20Testing%20of%20NLP%20Models%20with%20CheckList.2020.acl-main.442.pdf

A Fine-grained Multimedia Knowledge Extraction System. https://github.com/52Pig/paper_reading/blob/master/A%20Fine-grained%20Multimedia%20Knowledge%20Extraction%20System.2020.acl-demos.11.pdf

A System to Support the Analysis. https://github.com/52Pig/paper_reading/blob/master/A%20System%20to%20Support%20the%20Analysis.2020.acl-demos.32.pdf

Climbing towards NLU On Meaning, Form, and Understanding in the Age of Data.2020.acl-main.463 https://github.com/52Pig/paper_reading/blob/master/Climbing%20towards%20NLU%20On%20Meaning%2C%20Form%2C%20and%20Understanding%20in%20the%20Age%20of%20Data.2020.acl-main.463.pdf

Don’t Stop Pretraining Adapt Language Models to Domains and Tasks.2020.acl-main.740 https://github.com/52Pig/paper_reading/blob/master/Don%E2%80%99t%20Stop%20Pretraining%20Adapt%20Language%20Models%20to%20Domains%20and%20Tasks.2020.acl-main.740.pdf

How Can We Accelerate Progress Towards Human-like Linguistic.2020.acl-main.465 https://github.com/52Pig/paper_reading/blob/master/How%20Can%20We%20Accelerate%20Progress%20Towards%20Human-like%20Linguistic.2020.acl-main.465.pdf

人群扩散-基于在线社交广告 https://github.com/52Pig/paper_reading/blob/master/Audience%20Expansion%20for%20Online%20Social%20Network%20Advertising.pdf

Mastering the game of Go without.nature24270 https://github.com/52Pig/paper_reading/blob/master/Mastering%20the%20game%20of%20Go%20without.nature24270.pdf