这是何红亮同学保存每日记录的地方,GitHub上的第一个目录。</p>
在这里,我将记录每天自己看到的有收获的文章。比如:数据挖掘、机器学习、深度学习、NLP、数据竞赛、Python进阶及有关数据及建模方面内容。
在数据科学的道路上,能够慢慢进步和成长!坚持每天进步一点点!</p>
如有小伙伴愿意在数据道路上结伴而行,可以关注微信公众号《通往数据自由之路》在后台留言。
<br>Keywords: data mining machine learning deep learning(数据挖掘,机器学习,深度学习,自然语言处理 NLP)
20181219:
RelSUE:知识图谱相关实体搜索 (WSDM’19 录用论文介绍)
http://ws.nju.edu.cn/blog/2018/11/wsdm19/
20181218:
谈谈我在自然语言处理入门的一些个人拙见(zenRRan)
https://mp.weixin.qq.com/s?__biz=MzI3ODgwODA2MA==&mid=2247484527&idx=1&sn=627204c50b0ef9c6587eece083da48ba
谈谈我在自然语言处理进阶上的一些个人拙见(zenRRan)
https://mp.weixin.qq.com/s?__biz=MzI3ODgwODA2MA==&mid=2247485946&idx=1&sn=43e1217f59e5ac61a74b38823b6b4fac
20181217:
【特征选择】feature-selector工具助你一臂之力
https://mp.weixin.qq.com/s?__biz=MzIxNjA2ODUzNg==&mid=2651436810&idx=1&sn=fed7aeac5fc8c32cd533ed0ba6fbcea1
20181216:
YOCSEF「知识图谱」专题探索班成功举办,五大高校、三大企业共话知识图谱理论与未来
https://mp.weixin.qq.com/s?__biz=MzI5NTIxNTg0OA==&mid=2247493585&idx=1&sn=996e479480d339f0dc32205399de86d4
20181215:
Facebook开源NLP建模框架PyText,从论文到产品部署只需数天
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650753854&idx=2&sn=7768f2f5826c5ed68c1593641605979e
20181214:
推荐系统中的注意力机制——阿里深度兴趣网络(DIN)
https://zhuanlan.zhihu.com/p/51623339
20181213:
近期知识图谱顶会论文推荐(附PaperWeekly解读)
https://zhuanlan.zhihu.com/p/52232316
20181212:
「回顾」金融知识图谱的应用与探索
https://mp.weixin.qq.com/s?__biz=MzU1NTMyOTI4Mw==&mid=2247486838&idx=1&sn=e9fd6cb79e0814ad8ac6f71d0cee8b9b
20181211:
「回顾」自然语言处理中的多任务学习
https://mp.weixin.qq.com/s?__biz=MzU1NTMyOTI4Mw==&mid=2247486994&idx=1&sn=cc5868a3af3aba562463f74c2bcf8b3a
20181210:
自然语言处理中的深度迁移学习——文本预训练
https://mp.weixin.qq.com/s?__biz=MzU2OTA0NzE2NA==&mid=2247504215&idx=1&sn=69e52af7d4b18e5f667c7ad031a8862e
20181209:
博客 | Tensorflow系列专题(十一):RNN的应用及注意力模型
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650674337&idx=2&sn=711b3e1bb2d36e6734f9f1b275dcd4c2
20181208:
复现+改造:Hierarchical BiLSTM CNN for document-level sentiment classification
https://zhuanlan.zhihu.com/p/35606434
20181207:
从Word Embedding到Bert模型—自然语言处理中的预训练技术发展史
https://mp.weixin.qq.com/s?__biz=MzIzNjc0MTMwMA==&mid=2247486205&idx=1&sn=cda3258406551762f1f6c6f326b0b421
20181206:
一文详解LDA主题模型
https://zhuanlan.zhihu.com/p/31470216
20181205:
用深度学习(CNN RNN Attention)解决大规模文本分类问题 - 综述和实践
https://zhuanlan.zhihu.com/p/25928551
20181204:
练手|常见30种NLP任务的练手项目
https://ai.yanxishe.com/page/blogDetail/9562
20181203:
一文了解自然语言处理神经史
https://mp.weixin.qq.com/s?__biz=MzI0NTE4NjA0OQ==&mid=2658359117&idx=1&sn=936d1d788666ab39039ffdf7e44711e0
20181202:
NeurIPS 2018提前看:生物学与学习算法
https://zhuanlan.zhihu.com/p/51370635
20181201:
七个不容易被发现的生成对抗网络(GAN)用例
https://zhuanlan.zhihu.com/p/48110854
20181130:
这些“秘密武器”,让你轻松跻身Kaggle前2%
https://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651665068&idx=1&sn=f596ff694223e48c835b993be7e752bc
AAAI 2019 | 半监督学习也能自动化?南大和第四范式提出Auto-SSL
https://mp.weixin.qq.com/s?__biz=MzAwMjM2Njg2Nw==&mid=2653146262&idx=1&sn=219a1336a55cae0858fffabb4bf7c91d
20181129:
美团餐饮娱乐知识图谱——美团大脑揭秘
https://mp.weixin.qq.com/s?__biz=MjM5NjQ5MTI5OA==&mid=2651749520&idx=1&sn=e905e4ecefca3be58b46f59dcfb4b202
20181128:
知识图谱论文整理(来自Pelhans)
https://github.com/Pelhans/paper_list/tree/master/knowledge_graph
20181127:
当知识图谱遇上文本智能处理,会擦出怎样的火花?
https://zhuanlan.zhihu.com/p/48191132
20181126:(注:今天外出上海 27日补)
如何在Python中把机器学习模型转成API
https://zhuanlan.zhihu.com/p/50290226
20181125:
机器学习专家:我在Kaggle竞赛排名前2%的秘诀
https://zhuanlan.zhihu.com/p/48927063
20181124:
DeepMind 重磅论文:通过奖励模型,让 AI 按照人类意图行事
https://zhuanlan.zhihu.com/p/50629150
20181123:
经典论文复现:基于标注策略的实体和关系联合抽取
https://zhuanlan.zhihu.com/p/50702914
20181122:
一个完整的端到端机器学习项目
https://github.com/DeqianBai/Your-first-machine-learning-Project---End-to-End-in-Python
20181121:
美团大脑 | 知识图谱的建模方法及其应用
https://blog.csdn.net/dQCFKyQDXYm3F8rB0/article/details/83218579
20181120:
pair2vec 基于复合词对嵌入加强跨句推理模型
https://zhuanlan.zhihu.com/p/49421365
20181119:
ICLR 2019 | 有效稳定对抗模型训练过程,伯克利提出变分判别器瓶颈
https://zhuanlan.zhihu.com/p/50318422
20181118:
最全NLP反作弊攻略,从马蜂窝注水事件说起
https://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651664645&idx=1&sn=606c575b412c233273e615a610b4775d
20181117:
南开大学提出最新边缘检测与图像分割算法,精度刷新记录(附开源地址)
https://zhuanlan.zhihu.com/p/48300590
原创 | 王欣:多维深耕打造数字化银行
https://mp.weixin.qq.com/s?__biz=MzI0Nzc3MTQyMw==&mid=2247487296&idx=1&sn=ae73bb76fdae12dc642ef59b2db5e3bf
20181116:
如何训练你的ResNet(一):复现baseline,将训练时间从6分钟缩短至297秒
https://zhuanlan.zhihu.com/p/49742612
【开源】最强NLP模型BERT的Pytorch实现,提供转换脚本
https://zhuanlan.zhihu.com/p/49469143
20181115:
基于深度迁移学习进行时间序列分类
https://zhuanlan.zhihu.com/p/49920651
20181114:
NIPS 2018:谷歌大脑提出简单、分布式概率编程,可用TPU大规模训练
https://zhuanlan.zhihu.com/p/49838701
20181113:
UC Berkeley提出变分判别器瓶颈,有效提高对抗学习平衡性
https://zhuanlan.zhihu.com/p/49473219
20181112:
手把手带你复现ICCV 2017经典论文—PyraNet
https://zhuanlan.zhihu.com/p/49088283
20181111:
大白话讲解卷积神经网络工作原理
https://zhuanlan.zhihu.com/p/49184702
强化学习从入门到放弃的资料
https://zhuanlan.zhihu.com/p/34918639
20181110:
WGAN-div:默默无闻的WGAN填坑者(附开源代码)
https://zhuanlan.zhihu.com/p/49046736
推荐系统遇上深度学习(二十四)--深度兴趣进化网络DIEN原理及实战!
https://zhuanlan.zhihu.com/p/48601882
20181109:
收下这16篇最新论文,周会基本不用愁
https://zhuanlan.zhihu.com/p/48926605
20181108:
TensorFlow系列专题(五):BP算法原理
https://zhuanlan.zhihu.com/p/48976225
NIPS 2018 | 南大周志华等人提出无组织恶意攻击检测算法UMA
https://zhuanlan.zhihu.com/p/48968937
20181107:
NIPS 2018 | 程序翻译新突破:UC伯克利提出树到树的程序翻译神经网络
https://zhuanlan.zhihu.com/p/48738352
20181106:
三年拿到斯坦福CS博士的创业者李纪为:AI如何赋能金融
https://zhuanlan.zhihu.com/p/48606594
20181105:
分享一篇比较全面的AutoML综述
https://zhuanlan.zhihu.com/p/48642938
腾讯PocketFlow中的模型压缩算法(一)剪枝篇
https://zhuanlan.zhihu.com/p/48420428
20181104:
收下这 16 篇最新论文,周会基本不用愁
https://mp.weixin.qq.com/s?__biz=MzIwMTc4ODE0Mw==&mid=2247492583&idx=1&sn=85ec5352079218745428d66ab8ee97d4
深度特征合成:自动化特征工程的运作机制
https://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651664000&idx=1&sn=ae6dda80df6d6278ae33b7bf7fbadcd2
20181103:
干货|饿了么推荐算法演进及在线学习实践
https://mp.weixin.qq.com/s?__biz=MzI3NTEyNTM4NQ==&mid=2649502705&idx=1&sn=7a7fbc3a39d9df0bbcc2bcb105aceb7b
20181102:
谷歌终于开源BERT代码:3 亿参数量,机器之心全面解读
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650751075&idx=2&sn=0a3ecd1af5f8549051760775e34db342
美团大脑:知识图谱的建模方法及其应用
https://mp.weixin.qq.com/s?__biz=MjM5NjQ5MTI5OA==&mid=2651749250&idx=1&sn=6f382206bed8d5e79999a8c957857ba5
20181101:
一文概览用于图像分割的CNN
https://zhuanlan.zhihu.com/p/48106193
20181031:
ACL 2018论文解读 | 基于排序**的弱监督关系抽取选种与降噪算法(张俊)
https://zhuanlan.zhihu.com/p/48049387
经典论文复现 | ICML 2017大热论文:Wasserstein GAN
https://zhuanlan.zhihu.com/p/48115362
20181030:
相关性≠因果:概率图模型和do-calculus
https://zhuanlan.zhihu.com/p/48054578
<EYD与机器学习>迁移学习基础
https://zhuanlan.zhihu.com/p/47991650
20181029:
NIPS 2018 | 行人重识别告别辅助姿势信息,商汤、中科大提出姿势无关的特征提取GAN
https://zhuanlan.zhihu.com/p/47954794
20181028:
经典论文复现 | InfoGAN:一种无监督生成方法
https://zhuanlan.zhihu.com/p/47850287
20181027:
使用深度学习打造智能聊天机器人
https://blog.csdn.net/malefactor/article/details/51901115
自己动手做聊天机器人 二十七-用深度学习来做自动问答的一般方法
http://www.shareditor.com/blogshow?blogId=104
20181026:
中文文本分类:你需要了解的10项关键内容
https://zhuanlan.zhihu.com/p/47761862
深度特征合成:自动化特征工程的运作机制
https://zhuanlan.zhihu.com/p/47746152
20181025:
EMNLP 2018 | 从对话生成和文本风格转化看文本生成技术
https://zhuanlan.zhihu.com/p/47622150
20181024:
深度学习用于NLP(分词)中的论文及代码集锦
https://mp.weixin.qq.com/s?__biz=MzI0MjU1NTk3NA==&mid=2247485359&idx=1&sn=d986bdfe878bd0037ff03d31112fb89d
20181023:
Kaggle竞赛「找盐」冠军:价值5万美元的第一名方案出炉
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650750516&idx=1&sn=e2d1e4c29a71c94827532d88ad5b59be
20181022:
自然语言处理中的语言模型预训练方法
https://zhuanlan.zhihu.com/p/47342053
20181021:
NLP应该如何学、如何教?斯坦福大学大牛Dan Jurafsky教授专访
https://mp.weixin.qq.com/s?__biz=MzU1NTUxNTM0Mg==&mid=2247489586&idx=2&sn=f8dc9511dcc0c1e13f0cf0fad79f0319
20181020:
第二届智慧**杯精品旅行服务成单预测 第二名比赛方案
https://github.com/Bifzivkar/Boutique-Travel-Services-Predict
20181019:
【专家解读】狂破11项记录,谷歌年度最强NLP论文BERT模型到底强在哪里?
https://zhuanlan.zhihu.com/p/46880276
20181018:
十大经典排序算法(动图演示,收藏好文)
https://mp.weixin.qq.com/s?__biz=MzU1NTUxNTM0Mg==&mid=2247489503&idx=1&sn=268936923992183fcefbb30f9698ada6
深度学习中的Normalization模型
https://mp.weixin.qq.com/s?__biz=MzU1NTUxNTM0Mg==&mid=2247489568&idx=3&sn=16c7b17f50e3dc8744debc228dc3954b
20181017:
经典论文复现:基于特征金字塔网络的人体姿态估计
https://zhuanlan.zhihu.com/p/46991726
20181016:
深度学习中的Lipschitz约束:泛化与生成模型
https://zhuanlan.zhihu.com/p/46924315
20181015:
论文笔记:图像数据增强之弹性形变(Elastic Distortions)
https://zhuanlan.zhihu.com/p/46833956
20181014:
基于深度学习的文字识别(3755个汉字)
https://www.cnblogs.com/skyfsm/p/8443107.html
20181013:
构建深度神经网络的一些实战建议-绝对的纯干货
https://zhuanlan.zhihu.com/p/46682252
20181012:
深度学习、机器学习与NLP的前世今生
https://zhuanlan.zhihu.com/p/41583326
20181011:
神经网络-标准神经网络(python3)
https://zhuanlan.zhihu.com/p/38065189
神经网络-卷积神经网络(python3)
https://zhuanlan.zhihu.com/p/38176963
20181010:
NIPS 2018 | MIT等提出NS-VQA:结合深度学习与符号推理的视觉问答
https://zhuanlan.zhihu.com/p/46392910
20181009:
从各种注意力机制窥探深度学习在NLP中的神威
https://zhuanlan.zhihu.com/p/46260179
20181008:
kaggle | talkingdata广告反欺诈 top 4% 总结
https://zhuanlan.zhihu.com/p/36945368
20181007:
【学界】第七章:Python代码之模块整合与主程序(知乎作者:小杨)
https://zhuanlan.zhihu.com/p/44357003
kaggle | Machine Learning for Insights Challenge(知乎作者:Young)
https://zhuanlan.zhihu.com/p/45898896
20181006:
【学界】第六章:Python代码之蚁群算法部分
https://zhuanlan.zhihu.com/p/46068608
ECCV2018 如何让RNN神经元拥有基础通用的注意力能力
https://zhuanlan.zhihu.com/p/43799786
20181005:
【学界】第五章:Python代码之模型与Q-Learning
https://zhuanlan.zhihu.com/p/46018470
神经网络-标准神经网络(python3)
20181004:
PCA & Autoencoders通俗解析
https://zhuanlan.zhihu.com/p/45737420
20181003:
每个Kaggle冠军的获胜法门:揭秘Python中的模型集成
https://zhuanlan.zhihu.com/p/32949396
20181002:
超真实AI假照片,BigGAN
https://zhuanlan.zhihu.com/p/45718568
最前沿:用模仿学习来学习增强学习
https://zhuanlan.zhihu.com/p/45845001
20181001:
文字的识别与定位:车牌识别(是真的车牌)和简单OCR为例
https://jizhi.im/blog/post/gpu-p5
20180930:
笔触可控的实时图像风格化迁移(ECCV2018)
https://zhuanlan.zhihu.com/p/39622999
20180929:
经典论文复现:基于深度学习的图像超分辨率重建
https://zhuanlan.zhihu.com/p/45627373
基于深度学习的行人重识别研究综述
https://zhuanlan.zhihu.com/p/31921944
20180928:
模型调优与模型融合(代码应用篇)
https://blog.csdn.net/qq_39422642/article/details/78566763
20180927:
<EYD与机器学习>:对抗攻击基础知识(十一)
https://zhuanlan.zhihu.com/p/45210478
20180926:
[天池智慧交通预测挑战赛]新手向的时间序列预测解决方案-前Top2%
https://blog.csdn.net/ppp8300885/article/details/77934822
20180925:
Kaggle实战——点击率预估
https://blog.csdn.net/chengcheng1394/article/details/78940565
20180924:
DataCastle-城市治理大数据应用创意方案赛 冠军代码
https://github.com/poteman/DataCastle-Urban-governance-competition
20180923:
CW-RNN 收益率时间序列回归
https://zhuanlan.zhihu.com/p/27104240
RNN最大化条件概率参数模型预测时间序列
https://zhuanlan.zhihu.com/p/26434409
NeuralGPU卷积与RNN结合超线性计算时间多因子时间序列预测
https://zhuanlan.zhihu.com/p/25757053
20180922:
MIT韩松团队:AutoML自动模型压缩,全面超越手工调参
https://zhuanlan.zhihu.com/p/44581463
20180921:
机器学习与深度学习常见面试题(上)
https://zhuanlan.zhihu.com/p/45091568
20180920:
【AI in 美团】深度学习在美团搜索广告排序的应用实践
https://zhuanlan.zhihu.com/p/37823302
20180919:
23种Pandas核心操作
https://zhuanlan.zhihu.com/p/43018099
新手必看:生成对抗网络的初学者入门指导
https://zhuanlan.zhihu.com/p/44917088
20180918:
变分自编码器VAE:一步到位的聚类方案
https://zhuanlan.zhihu.com/p/44808113
20180917:
200种最佳机器学习教程汇总【史上最全】(阿里云云栖社区)
https://zhuanlan.zhihu.com/p/44710585
20180916:
CNN入门实战:我如何把准确率从86% 提高到99%(中)
https://zhuanlan.zhihu.com/p/44219807
20180915:
R语言之系统聚类(层次)分析之图谱形式完整版
https://blog.csdn.net/hhtnan/article/details/79281721
20180914:
【用心整理的论文-代码实现列表(w/stars)】“Papers with code. Sorted by stars. Updated weekly.” by Zaur Fataliyev GitHub
https://github.com/zziz/pwc
20180913:
无需写代码!谷歌推出机器学习模型分析神器,代号What-If
https://zhuanlan.zhihu.com/p/44357475
调参工要凉?微软重磅开源AutoML工具包NNI
https://mp.weixin.qq.com/s?__biz=MzU1NDA4NjU2MA==&mid=2247492720&idx=1&sn=e6172b6ae7276845d37118a9ebe34527
20180912:
CycleGAN原理以及代码全解析
https://zhuanlan.zhihu.com/p/37198143
20180911:
用ARIMA构建时间序列模型(附Python 和R代码)
https://zhuanlan.zhihu.com/p/43771241
20180910:
谷歌带来GAN入门神器:浏览器上运行的可视化工具GAN Lab
https://www.toutiao.com/a6599505914636534280
20180909:
【ECCV2018】24篇论文代码实现
https://mp.weixin.qq.com/s?__biz=MzU2OTA0NzE2NA==&mid=2247496997&idx=1&sn=af765489db4d09e18a1017035f6f3b8a
20180908:
ICML2018生成对抗网络论文评述
https://zhuanlan.zhihu.com/p/43874842
20180907:
【DL笔记8】如果你愿意一层一层剥开CNN的心
https://zhuanlan.zhihu.com/p/42904109
深度学习贝叶斯,这是一份密集的6天速成课程(视频与PPT)
https://zhuanlan.zhihu.com/p/43973060
20180906:
CNN入门实战:我如何把准确率从86% 提高到99%(上)
https://zhuanlan.zhihu.com/p/43679939
20180905:
Kaggle前1%经验:ML竞赛中常被忽视的特征工程技术
https://zhuanlan.zhihu.com/p/43691981
20180904:
12种降维方法终极指南(含Python代码)
https://zhuanlan.zhihu.com/p/43225794
均已拿offer|19校招阿里腾讯华为美团算法岗面经
https://zhuanlan.zhihu.com/p/43638109
20180903:
首战拿银总结--无套路的谈一谈kaggle比赛
https://zhuanlan.zhihu.com/p/43472504
20180902:
深度学习中的Normalization模型
https://zhuanlan.zhihu.com/p/43200897
20180901:
CCKS2018最佳论文:南京大学提出DSKG—知识图谱补全
https://zhuanlan.zhihu.com/p/42887475
20180831:
十大深度学习热门论文(2018年版)
https://zhuanlan.zhihu.com/p/42509495
20180830:
KDD Cup 2018 冠军「 first floor to eat latiao」:为什么取这个队名?因为大家都爱辣条
https://mp.weixin.qq.com/s?__biz=MzI5NTIxNTg0OA==&mid=2247492360&idx=1&sn=a91cd339f1f35135e37c8d70983d780c
20180829:
机器学习实战|GBDT Xgboost LightGBM对比
https://zhuanlan.zhihu.com/p/43181861
20180828:
算法/NLP/深度学习/机器学习面试笔记
https://github.com/imhuay/Algorithm_Interview_Notes-Chinese
20180827:
计算:XGBoost背后的数学原理
https://zhuanlan.zhihu.com/p/42489958
20180826:
无法进入的领域: 树模型的失落
https://mp.weixin.qq.com/s?__biz=MzU2NjY2NzkzMA==&mid=2247483669&idx=1&sn=716dfb17ec29546ba3c53135918a9616
20180825:
腾讯广告算法赛Top4—SDD模型1—FM_keras
https://blog.csdn.net/ML_SDD/article/details/81702142
20180824:
「特征工程」与「表示学习」
https://zhuanlan.zhihu.com/p/41521695
比RNN快136倍!上交大提出SRNN,能做并行计算
https://zhuanlan.zhihu.com/p/42242327
20180823:
吴恩达deeplearning.ai学完总结(来自知乎毛帅)
https://zhuanlan.zhihu.com/p/41161643
当前训练神经网络最快的方式:AdamW优化算法+超级收敛
https://zhuanlan.zhihu.com/p/38945390
20180822:
利用层次聚类算法进行基于基站定位数据的商圈分析
https://mp.weixin.qq.com/s?__biz=MzAxMjUyNDQ5OA==&mid=2653557145&idx=1&sn=194fde7fce9ec8745a281fbda59d1b05
20180821:
2018腾讯广告算法大赛第三名开源代码 Tencent 2018 Advertisement Competition NO.3
https://github.com/DiligentPanda/Tencent_Ads_Algo_2018
20180820:
2018/2019/校招/春招/秋招/自然语言处理(NLP)/深度学习(Deep Learning)/机器学习(Machine Learning)/C/C++/Python/面试笔记
https://github.com/imhuay/Interview_Notes-Chinese
20180819:
44页PPT详解R语言、时间序列与机器学习
https://zhuanlan.zhihu.com/p/42285498
手把手教你完成一个数据科学小项目(5):省份提取与可视化
https://zhuanlan.zhihu.com/p/42406348
20180818:
纯技术干货|深度学习文本分类在支付宝安全风控上的应用
https://zhuanlan.zhihu.com/p/42215024
20180817:
NLP通用模型诞生?一个模型搞定十大自然语言常见任务
https://mp.weixin.qq.com/s?__biz=MzIzNzA4NDk3Nw==&mid=2457736381&idx=1&sn=d8981b0e6ae24e392f392973e146ddc9
20180816:
美团O2O广告营销中的机器学习技术
https://mp.weixin.qq.com/s?__biz=MjM5NjQ5MTI5OA==&mid=2651748603&idx=1&sn=b09362d81941732e7d9c69f309a9e2e2
20180815:
网络表示学习综述:一文理解Network Embedding
https://zhuanlan.zhihu.com/p/42022918
20180814:
算法/NLP/深度学习/机器学习面试笔记
https://github.com/imhuay/Interview_Notes-Chinese
20180813:
入门 | 献给新手的深度学习综述
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650746791&idx=2&sn=6a9c81461f3ac4588e8d5de2d2c1ded9
聊聊FFM在CTR预估中的应用
https://mp.weixin.qq.com/s?__biz=MzI0MjU1NTk3NA==&mid=2247484627&idx=1&sn=38bc163934af6002de928689f9abb74e
20180812:
自然语言处理是如何工作的?一步步教你构建 NLP 流水线
https://zhuanlan.zhihu.com/p/41850756
20180811:
Julia 1.0 正式发布,这是新出炉的一份简单中文教程
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650746902&idx=1&sn=d3f55a745cc0575e888e060da19b46be
20180810:
GANs有嘻哈:一次学完10个GANs明星模型(附视频)
https://zhuanlan.zhihu.com/p/41722006
《纯干货》2018-2019年国际AI会议最全信息整理分享
https://zhuanlan.zhihu.com/p/41695243
20180809:
推荐系统遇上深度学习(十八)-探秘阿里深度兴趣网络浅析及实现
https://zhuanlan.zhihu.com/p/38875771
20180808:
称霸Kaggle的十大深度学习技巧
https://zhuanlan.zhihu.com/p/41379279
47页PPT,用户画像架构、指标标签、ETL性能及案例一站通
https://zhuanlan.zhihu.com/p/41547953
20180807:
对比学习用 Keras 搭建 CNN RNN 等常用神经网络
https://www.jianshu.com/p/9efae7a20493
20180806:
【干货合集】通俗理解kaggle比赛大杀器xgboost
https://zhuanlan.zhihu.com/p/41417638
【干货合集】一文读懂特征工程
https://zhuanlan.zhihu.com/p/41156662
20180805:
在机器学习的项目中,特征是如何被找出来的?
https://www.zhihu.com/question/41659582/answer/434804910
20180804:
CTR预估算法演进
https://zhuanlan.zhihu.com/p/39530235
20180803:
CTR预估专栏 | Facebook经典模型LR+GBDT理论与实践
https://mp.weixin.qq.com/s?__biz=MzU1NDA4NjU2MA==&mid=2247492002&idx=2&sn=625607e9bc58ff0d8aaa162494f82b64
20180802:
CVPR 2018:基于域适应弱监督学习的目标检测
https://zhuanlan.zhihu.com/p/41126114
20180801:
机器之心专访杨强教授:联邦迁移学习与金融领域的AI落地
https://zhuanlan.zhihu.com/p/41052548
20180731:
阿里音乐流行趋势预测大赛,赛后总结
https://www.cnblogs.com/xuancaoyy/p/5822755.html
20180730:
终极指南:构建用于检测汽车损坏的Mask R-CNN模型
https://zhuanlan.zhihu.com/p/40896926
20180729:
最前沿:Meta Learning前沿进展扫描
https://zhuanlan.zhihu.com/p/40600485
第二届腾讯广告算法大赛总结(Rank 9)
https://zhuanlan.zhihu.com/p/38499275
20180728:
「干货」计算机视觉视频理解领域的经典方法和最新成果
https://www.toutiao.com/a6560486800022831629
20180727:
基于 Keras 用深度学习预测时间序列
https://www.cnblogs.com/xuruilong100/p/8413683.html
20180726:
提供免费计算资源,开发平台AIStudio零门槛实现AI能力
https://zhuanlan.zhihu.com/p/40643178
20180725:
2018腾讯广告算法大赛总结/0.772229/Rank11
https://zhuanlan.zhihu.com/p/38034501
特征选择
https://zhuanlan.zhihu.com/p/32749489
20180724:
数据挖掘和机器学习的面试问题
https://zhuanlan.zhihu.com/p/40559938
20180723:
自己动手打造深度学习服务器
https://zhuanlan.zhihu.com/p/40365673
20180722:
深度学习在CTR预估中的应用
https://zhuanlan.zhihu.com/p/35484389
20180721:
一文读懂机器学习大杀器XGBoost原理
https://zhuanlan.zhihu.com/p/40129825
20180720:
推荐系统遇上深度学习(十六)--详解推荐系统中的常用评测指标
https://zhuanlan.zhihu.com/p/38875570
20180719:
ICML 2018 | 模型层面的对偶学习
https://zhuanlan.zhihu.com/p/40115270
20180718:
反欺诈(Fraud Detection)中所用到的机器学习模型有哪些?
https://www.zhihu.com/question/30508773
20180717:
Yann LeCun:学习世界模型,通向AI的下一步
https://zhuanlan.zhihu.com/p/39928037
从FM推演各深度CTR预估模型(附代码)
https://blog.csdn.net/longxinchen_ml/article/details/81031736
20180716:
教程 | 比Python快100倍,利用spaCy和Cython实现高速NLP项目
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650745273&idx=2&sn=bb811e34549e917a7734485e782a1309
20180715:
Kaggle竞赛-房价预测(House Prices)小结(作者:Npein小林)
https://zhuanlan.zhihu.com/p/39429689
20180714:
AI与区块链结合如果结合?这里有几个案例与一点方向思考
https://zhuanlan.zhihu.com/p/39714341
特征金字塔网络FPN
https://zhuanlan.zhihu.com/p/39185919
20180713:
换脸效果媲美GAN!一文解析OpenAI流生成模型Glow
https://zhuanlan.zhihu.com/p/39676312
[阿里]独家 | 一文读懂复杂网络(应用、模型和研究历史)
https://zhuanlan.zhihu.com/p/35705787
20180712:
基于django开发的一个轻量级办公平台
https://zhuanlan.zhihu.com/p/38340557
20180711:
时间序列异常点检测算法
https://zhuanlan.zhihu.com/p/39453139
生成式对抗网络模型综述
https://zhuanlan.zhihu.com/p/39398823
20180710:
资源|Kaggle数据科学项目索引表,10大类93项,更新中
https://zhuanlan.zhihu.com/p/39422753
下一个GAN?OpenAI提出可逆生成模型Glow
https://zhuanlan.zhihu.com/p/39407274
20180709:
NLP领域的ImageNet时代:词嵌入已死,语言模型当立
https://zhuanlan.zhihu.com/p/39326211
深度学习&CV方向|腾讯优图、依图、百度面经分享
https://zhuanlan.zhihu.com/p/39334253
20180708:
入门Python神经机器翻译,这是一篇非常精简的实战指南
https://zhuanlan.zhihu.com/p/39265628
20180707:
如何保证算法公正性?ICML 2018两篇获奖论文解读
https://zhuanlan.zhihu.com/p/39197413
20180706:
CNN也能用于NLP任务,一文简述文本分类任务的7个模型
https://zhuanlan.zhihu.com/p/39054002
2018腾讯社交算法大赛 Tencent2018_Lookalike_Rank7th
https://github.com/guoday/Tencent2018_Lookalike_Rank7th
20180705:
机器学习算法地图
https://zhuanlan.zhihu.com/p/39020459
【HFT系列】基于机器学习的动态高频限价订单簿框架(Tick数据)
https://mp.weixin.qq.com/s?__biz=MzAxNTc0Mjg0Mg==&mid=2653288278&idx=1&sn=73c6749fa89384391031c78a55768681
20180704:
如何在第一次天池比赛中进入Top 5%(一)来自[字节]
https://zhuanlan.zhihu.com/p/23845169
如何在 Kaggle 首战中进入前 10% 来自[Wille]
https://dnc1994.com/2016/04/rank-10-percent-in-first-kaggle-competition/
20180703:
当前训练神经网络最快的方式:AdamW优化算法+超级收敛
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650744672&idx=1&sn=969f9f655b6fa97baa79db079ec8eb18
20180702:
自然语言十项全能:转化为问答的多任务学习
https://zhuanlan.zhihu.com/p/38467146
20180701:
机器学习算法面试小结
https://zhuanlan.zhihu.com/p/29677765
20180630:
attention模型方法综述
https://zhuanlan.zhihu.com/p/37835894
20180629:
【AI in 美团】深度学习在文本领域的应用
https://mp.weixin.qq.com/s?__biz=MjM5NjQ5MTI5OA==&mid=2651748157&idx=2&sn=1d5431961d4db29682b57daf169e3a81
【AI in 美团】深度学习在OCR中的应用
https://mp.weixin.qq.com/s?__biz=MjM5NjQ5MTI5OA==&mid=2651748259&idx=2&sn=2e67dea28376f04dabcb6b5dd787a6e7
20180628:
推荐系统遇上深度学习(九)--评价指标AUC原理及实践
https://zhuanlan.zhihu.com/p/37522326
20180627:
【重磅】基于深度学习的目标检测算法综述
https://zhuanlan.zhihu.com/p/36088972
20180626:
Meta Learnjng:遗传算法与深度学习的火花(附github代码)
https://mp.weixin.qq.com/s?__biz=MzI5MDUyMDIxNA==&mid=2247485732&idx=1&sn=902207ddfe145ab1a1b1900d8512b819
推荐系统遇上深度学习(十八)--探秘阿里之深度兴趣网络(DIN)浅析及实现
https://mp.weixin.qq.com/s?__biz=MzI1MzY0MzE4Mg==&mid=2247484002&idx=1&sn=e88a5eeb1ead6f1650dbe1cbe5c1fd0f
20180625:
机器学习论文+代码大全,这个网站请收藏好(http://www.gitxiv.com/)
https://zhuanlan.zhihu.com/p/38468141
2018腾讯广告算法大赛总结(Rank6)-模型篇
https://zhuanlan.zhihu.com/p/38443751
20180624:
用Python实现模拟登录正方教务系统抢课
https://zhuanlan.zhihu.com/p/38057708
一文带你了解Python集合与基本的集合运算
https://zhuanlan.zhihu.com/p/38442300
20180623:
Kaggle竞赛方案分享:如何分辨杂草和植物幼苗
https://zhuanlan.zhihu.com/p/38359300
20180622:
Feature Tools:可自动构造机器学习特征的Python库
https://zhuanlan.zhihu.com/p/38349856
20180621:
推荐系统遇上深度学习(五)--Deep&Cross Network模型理论和实践
https://zhuanlan.zhihu.com/p/37522234
20180620:
TCRP风控管理师知识点:信贷&评分卡模型&授信
https://zhuanlan.zhihu.com/p/38273688
20180619:
推荐系统遇上深度学习(六)--PNN模型理论和实践
https://zhuanlan.zhihu.com/p/37522266
20180618:
2018IJCAI-阿里妈妈广告搜索转化预测大赛/腾讯广告算法大赛总结
https://zhuanlan.zhihu.com/p/38202468
20180617:
文本检测之PixelLink
https://zhuanlan.zhihu.com/p/38171172
20180616:
粒子群优化算法(PSO)之基于离散化的特征选择(FS)(三)
https://zhuanlan.zhihu.com/p/37139588
20180615:
Python爬虫基础练习(十) 验证码爬取+识别
https://zhuanlan.zhihu.com/p/38057424
数据挖掘中常见的『异常检测』算法有哪些?
https://www.zhihu.com/question/280696035/answer/417665007
20180614:
推荐系统遇上深度学习(五)--Deep&Cross Network模型理论和实践
https://zhuanlan.zhihu.com/p/37522234
20180613:
模型评估——定量分析预测的质量
https://blog.csdn.net/hustqb/article/details/77922031
20180612:
对抗攻击基础知识
https://zhuanlan.zhihu.com/p/37922148
20180611:
用人工蜂群算法求解k-分区聚类问题
https://zhuanlan.zhihu.com/p/37960107
20180610:
深度学习模型的简单优化技巧
https://zhuanlan.zhihu.com/p/37918855
20180609:
一文读懂如何用LSA、PSLA、LDA和lda2vec进行主题建模
https://zhuanlan.zhihu.com/p/37873878
20180608:
常用测试集带来过拟合?你真的能控制自己不根据测试集调参吗
https://zhuanlan.zhihu.com/p/37799067
20180607:
说说表示学习和mGBDT(multi-layered GBDT)
https://zhuanlan.zhihu.com/p/37712433
推荐系统遇上深度学习(四)--多值离散特征的embedding解决方案
https://zhuanlan.zhihu.com/p/37522179
20180606:
DenseNet:比ResNet更优的CNN模型
https://zhuanlan.zhihu.com/p/37189203
16岁高中生夺冠Kaggle地标检索挑战赛!而且竟然是Kaggle老兵
https://zhuanlan.zhihu.com/p/37522227
20180605:
深度森林第三弹:周志华组提出可做表征学习的多层梯度提升决策树
https://zhuanlan.zhihu.com/p/37727525?group_id=987401007302787072
20180604:
CVPR 2018 | 牛津大学&Emotech首次严谨评估语义分割模型对对抗攻击的鲁棒性
https://zhuanlan.zhihu.com/p/37662586
机器学习之环境配置(1)—win10/ubuntu18.04双系统+CUDA9.0+CUDNN7.0.5+tensorflow1.8
https://zhuanlan.zhihu.com/p/37651456
20180603:
走近流行强化学习算法:最优Q-Learning
https://zhuanlan.zhihu.com/p/37620639
20180602:
Ubuntu16.04+Ananconda3+CUDA+CUDNN+Tensorflow-gpu配置教程
https://zhuanlan.zhihu.com/p/37569310
CVPR 2018 | 旷视科技人体姿态估计冠军论文——级联金字塔网络CPN
https://zhuanlan.zhihu.com/p/37582402
20180601:
CVPR 2018:阿里提出新零样本学习方法,有效解决偏置问题
https://mp.weixin.qq.com/s?__biz=MzU1NTUxNTM0Mg==&mid=2247488722&idx=1&sn=78ff581bd26c28f2721688bee9c0a39b
仅使用NumPy完成卷积神经网络CNN的搭建(附Python代码)
https://mp.weixin.qq.com/s?__biz=MzI0NTE4NjA0OQ==&mid=2658357398&idx=1&sn=1c4fe04f915cb14fb222563469f1b8ff
20180531:
如何靠数据分析“上位”?——一位银行业务分析专家的实践
https://zhuanlan.zhihu.com/p/37483791
20180530:
进化计算在深度学习中的应用 | 附多篇论文解读
https://zhuanlan.zhihu.com/p/37493469
20180529:
你“听”过这些经典排序算法吗?
https://zhuanlan.zhihu.com/p/37430943
20180528:
深度 | 蚂蚁金服DASFAA论文带你深入了解GBDT模型
https://mp.weixin.qq.com/s?__biz=MzI0Nzc3MTQyMw==&mid=2247484910&idx=1&sn=f517eb0226e3f75a52497fd7682a2d1a
20180527:
一人署名!清华特奖得主、姚班本科生乔明达ICML 2018论文提出协作学习的鲁棒性方法
https://zhuanlan.zhihu.com/p/37351944
20180526:
非线性特征提取和模型堆叠
https://zhuanlan.zhihu.com/p/37313922
20180525:
你想要的Python面试都在这里了【315+道题】
https://zhuanlan.zhihu.com/p/37215216
Python进行微信公众号开发
https://zhuanlan.zhihu.com/p/37229184
20180524:
兜哥出品 <一本开源的NLP入门书籍>
https://github.com/duoergun0729/nlp
小姐姐的游戏运营面经分享:腾讯+阿里
https://zhuanlan.zhihu.com/p/37142126
20180523:
一文简述多种无监督聚类算法的Python实现
https://zhuanlan.zhihu.com/p/37152132
推荐系统遇上深度学习(十)--GBDT+LR融合方案实战
https://mp.weixin.qq.com/s?__biz=MzI1MzY0MzE4Mg==&mid=2247483933&idx=1&sn=514472ac10955647fdead331275e4083
20180522:
马尔可夫链蒙特卡罗算法(MCMC)及python实现
https://zhuanlan.zhihu.com/p/37121528
20180521:
循环神经网络综述 -语音识别与自然语言处理的利器(上篇)
https://zhuanlan.zhihu.com/p/37036777
循环神经网络综述 -语音识别与自然语言处理的利器(下篇)
https://zhuanlan.zhihu.com/p/37102550
20180520:
常用机器学习常用算法优点及缺点总结
https://zhuanlan.zhihu.com/p/36928215
20180519:
一文带你理解Q-Learning的搜索策略
https://zhuanlan.zhihu.com/p/37048004
20180518:
为什么做机器学习的很少使用假设检验?(微调的回答)
https://www.zhihu.com/question/55420602/answer/394577531
20180517:
Kaggle 恶意评论(toxic comment classification)分类 top 1 %方案
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670619&idx=1&sn=ea66bc6e42a1ce299306f72df88a0bfb
kaggle | talkingdata广告反欺诈 top 4% 总结
https://zhuanlan.zhihu.com/p/36945368
机器学习小白看过来,带你全面了解分类器评价指标
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670816&idx=1&sn=b5effa31f9c45d2df0eb2d5f7a4be0a5
理解梯度下降法
https://zhuanlan.zhihu.com/p/36902908
20180516:
Must Know! 数据科学家们必须知道的 5 种聚类算法
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670292&idx=1&sn=2bfd99e74b6e4f656ea6f84a47a51ae7
如何构建商品定价模型?Mercari Price Suggestion Challenge 最佳方案出炉
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670353&idx=1&sn=539b46f43496b8f9cdb05dd04b88d37c
Kaggle 商品销量预测季军方案出炉,应对时间序列问题有何妙招
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670404&idx=1&sn=f751b8835f4e37173551bfd65b43d193
20180515:
[天池阿里妈妈搜索广告转化预测]参赛分享|附R代码
https://zhuanlan.zhihu.com/p/36784736
【天池-算法入门】sklearn入门系列三:数据预处理、模型评估、模型融合
https://tianchi.aliyun.com/forum/videoStream.html?spm=5176.11409386.0.0.328b311f647pfD&postsId=4962#postsId=4962
20180514:
Kaggle 大神 Eureka 的高手进阶之路
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670135&idx=1&sn=8b5fe7b489772bbf5575b4d94b7be08d
20180513:
Kaggle冠军冲顶经验分享:怎样11步搞定机器学习竞赛?
https://mp.weixin.qq.com/s?__biz=MzIzNjc1NzUzMw==&mid=2247498296&idx=6&sn=a2e6d0e539299eee2891a93f4c97b13c
20180512:
2018春季校园招聘笔经面经合集:数据分析岗
https://zhuanlan.zhihu.com/p/36689662
20180511:
十面阿里,七面头条,六个Offer,春招结束
https://zhuanlan.zhihu.com/p/36636356
如何通过梯度检验帮助实现反向传播
https://zhuanlan.zhihu.com/p/36648831
用户画像之标签聚类
https://zhuanlan.zhihu.com/p/29002395
20180510:
Sklearn 与 TensorFlow 机器学习实用指南(书)
https://github.com/apachecn/hands_on_Ml_with_Sklearn_and_TF/tree/dev/docs
20180509:
「融合」机器学习模型:一种提升预测能力的方法
https://zhuanlan.zhihu.com/p/33589222
20180508:
kaggle Talking Data 广告欺诈检测竞赛 top 1%方案分享
https://zhuanlan.zhihu.com/p/36580283
20180507:
面试必备 | 排序算法的Python实现
https://zhuanlan.zhihu.com/p/36419582
20180506:
2018年「机器学习/数据挖掘/人工智能」会议汇总(持续更新)
https://zhuanlan.zhihu.com/p/32974603
20180505:
机器学习中如何融合多种特征或分类方法?(微调的回答)
https://www.zhihu.com/question/24598447/answer/383717513
20180504:
sklearn入门系列三:数据预处理、模型评估、模型融合(主讲人BRYAN)
https://tianchi.aliyun.com/forum/videoStream.html?spm=5176.11409386.0.0.328b311f647pfD&postsId=4962#postsId=4962
20180503:
你的论文能否中顶会?这篇分析同行评审结果的论文可帮助你
https://zhuanlan.zhihu.com/p/36324468
拨开深度学习的迷雾:训练一个性能优秀的深度模型
https://zhuanlan.zhihu.com/p/36347673
2018腾讯广告算法大赛FFM baseline 0.731
https://zhuanlan.zhihu.com/p/36302396
20180502:
如何写一篇科技论文
https://zhuanlan.zhihu.com/p/34761704
20180501:
面试笔试整理3:深度学习机器学习面试问题准备(必会)
https://blog.csdn.net/woaidapaopao/article/details/77806273
20180430:
产品经理如何进行竞品分析?
https://www.zhihu.com/question/23601989/answer/376642573
生成对抗网络(GAN)
https://zhuanlan.zhihu.com/p/34287744
20180429:
使用sklearn做单机特征工程
http://www.cnblogs.com/jasonfreak/p/5448385.html
特征选择 (feature_selection)
https://www.cnblogs.com/stevenlk/p/6543628.html
机器学习之特征选择
https://www.cnblogs.com/nolonely/p/6435083.html
20180428:
SciHub桌面版官方发布页面暨Sci-Hub最新可用网址(20180427更新)
https://zhuanlan.zhihu.com/p/31809890
20180427:
这些资源你肯定需要!超全的GAN PyTorch+Keras实现集合
https://zhuanlan.zhihu.com/p/36058049
20180426:
主流CTR预估模型的演化及对比
https://zhuanlan.zhihu.com/p/35465875
20180425:
【干货】京东金融信贷需求预测竞赛,我曾经这样的思考过
https://mp.weixin.qq.com/s?__biz=MzI5ODQxMTk5MQ==&mid=2247484917&idx=1&sn=16cf9b89e69b781c978c9bce3a2870af
20180424:
如何通过“统计学检验”来对比机器学习算法性能
https://www.zhihu.com/question/27306416/answer/372241948
感动!有人将吴恩达的视频课程做成了文字版
https://mp.weixin.qq.com/s?__biz=MzAwNDI4ODcxNA==&mid=2652247185&idx=1&sn=438cb51af718432cb44b3581ce37ab75
20180423:
数据不足如何用深度学习 by Mike Bernico
https://www.bilibili.com/video/av22445154/?redirectFrom=h5
深度学习领域引用量前10篇论文(附下载地址)
https://zhuanlan.zhihu.com/p/36017938
20180422:
如何用Python批量提取PDF文本内容?
https://mp.weixin.qq.com/s?__biz=MzIyODI1MzYyNA==&mid=2653540360&idx=1&sn=efe357c982f934f9c29e03ec660f2d45
如何用Python和机器学习训练中文文本情感分类模型?
https://mp.weixin.qq.com/s?__biz=MzIyODI1MzYyNA==&mid=2653540357&idx=1&sn=72ccd564cc109f3f0b831af6a62e48c9
20180421:
如何准备机器学习工程师的面试?
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650667647&idx=1&sn=142c2aa04e2af0f700a0c8bce08c1de0
数据挖掘竞赛的套路就在这里了,看完本文全明白!
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650669620&idx=2&sn=158af9884aa6ef5d7a4317d3fff2ad09
20180420:
如何用自组织映射 (SOM) 解决旅行商问题 (TSP)
https://zhuanlan.zhihu.com/p/34121865
数据分析秘籍在这里:Kaggle 六大比赛最全面解析(上)
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670710&idx=1&sn=d75a077506c72eb5ba5d59c063cc0dda
数据分析秘籍在这里:Kaggle 六大比赛最全面解析(下)
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670719&idx=2&sn=27610e82a69ad97320afb754065b2a1f
20180419:
【回顾】从传统算法走向神经网络(视频讲解)
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650668271&idx=3&sn=da64ffd251aea27935242998f47f7c92
神经网络中容易被忽视的基础知识
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650667791&idx=1&sn=6f5fb4786ab8d8050827dffa77c3a012
20180418:
深度学习在文本分类中的应用
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650669100&idx=1&sn=1e82bd46dd189f8124a124ea50e89431
Kaggle Titanic 生存预测比赛超完整笔记(上)
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650668442&idx=1&sn=c26382573e26c52c27355917019b758a
Kaggle Titanic 生存预测比赛超完整笔记(中)
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650668479&idx=1&sn=cd0f1df7f7b6d0c522d8db19455afb40
Kaggle Titanic 生存预测比赛超完整笔记(下)
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650668486&idx=1&sn=e8149bb902d7692ea7dcd7721b136ad7
20180417:
Google Brain 团队 Hugo Larochelle 教授「Neural Networks」课程合辑!
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650669909&idx=1&sn=bfeb7c1bbf97e4ad4dcb027ccdd4148a
代码+实战:TensorFlow Estimator of Deep CTR —— DeepFM/NFM/AFM/FNN/PNN
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650669888&idx=1&sn=5fa4955db5fdb0a6f3b6c8e5893cf7df
视频 | 从原理到代码,Hinton 的胶囊网络该怎么用?(附代码)
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670011&idx=3&sn=273e2f5e33cc0994c46496b894e82aa6
从 Encoder 到 Decoder 实现 Seq2Seq 模型
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650669840&idx=1&sn=15e09738a64e1f8054e1db5b4f9360b4
20180416:
用 Keras 搭建 GAN:图像去模糊中的应用(附代码)
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670680&idx=1&sn=6dd0f7e1fc8b041ea590d615531053be
阿里巴巴 WSDM Cup 2018 夺得第二名,获奖论文全解读
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650670113&idx=1&sn=38670db51afc04464f410761ba6f50cc
Github 项目推荐 | GAN 的 Keras 实现案例集合 —— Keras-GAN
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650669978&idx=3&sn=d7ee5d09bee8731fb1552791ee19f46b
混合整数规划/离散优化的精确算法--分支定界法及优化求解器
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650669938&idx=1&sn=25ec13864509192e4b18856b234ccb40
20180415:
【爬虫军火库】如何优雅地复制请求头
https://mp.weixin.qq.com/s?__biz=MzI0OTEwOTk5MA==&mid=2648498823&idx=1&sn=c7833b78e7ff30ab329c4b5b4c6c4462
20180414:
ICASSP 2018 | 阿里巴巴论文提出针对影视作品的语音情感识别信息融合框架
https://zhuanlan.zhihu.com/p/35532054
20180413:
Python数据挖掘与机器学习,快速掌握聚类算法和关联分析
https://mp.weixin.qq.com/s?__biz=MzI1NDE1MDk2Mg==&mid=2247485013&idx=1&sn=6190c9dd7479bd2a2e97bbf27b3ea85c
20180412:
【Kaggle金融时间预测相关比赛整理】Dataset, Model and Winner's Solution
https://zhuanlan.zhihu.com/p/35363444
20180411:
手把手:用Python搭建一个加密货币交易模拟器,不用投钱就能玩
https://zhuanlan.zhihu.com/p/35507218
20180410:
深度学习在CTR预估中的应用
https://zhuanlan.zhihu.com/p/35484389
20180409:
连续特征的离散化:在什么情况下将连续的特征离散化之后可以获得更好的效果?
https://www.zhihu.com/question/31989952
20180408:
用户在线广告点击行为预测的深度学习模型
https://blog.csdn.net/xiewenbo/article/details/52038493
20180407:
DataFrame数据格式转换成FFM格式
https://github.com/luohangtong/DataFrame2FFM
Pandas Dataframe to libsvm format:有什么方便的方法吗?
https://www.zhihu.com/question/39038261/answer/108336665
20180406:
阿里新突破!自主创新的下一代匹配&推荐技术:任意深度学习+树状全库检索
https://yq.aliyun.com/articles/572353
新型RNN:将层内神经元相互独立以提高长程记忆 | CVPR 2018论文解读
https://zhuanlan.zhihu.com/p/35328410
20180405:
A curated list of resources for Chinese NLP 中文自然语言处理相关资料
https://github.com/crownpku/Awesome-Chinese-NLP
20180404:
【NSR特别专题】香港科技大学杨强:多任务学习概述
https://mp.weixin.qq.com/s?__biz=MzAxNDcxMzMxMw==&mid=2652667323&idx=1&sn=c4a0c93ea60334febc9ad26f3800e942
如何走近深度学习人脸识别?你需要这篇超长综述 | 附开源代码
https://zhuanlan.zhihu.com/p/35295839
20180403:
卷积的发展历程,原理和基于 TensorFlow 的实现
https://zhuanlan.zhihu.com/p/35226876
20180402:
通过kaggle比赛学习机器学习文本分类方法
https://zhuanlan.zhihu.com/p/34899693
20180401:
超快速!10分钟入门Keras指南
https://zhuanlan.zhihu.com/p/34461956
使用Keras进行深度学习:(二)CNN讲解及实践
https://zhuanlan.zhihu.com/p/34822375
深度学习--Lenet-5基本结构及keras实现
https://zhuanlan.zhihu.com/p/31612931
20180331:
深入FFM原理与实践
https://tech.meituan.com/deep-understanding-of-ffm-principles-and-practices.html
20180330:
人工智能最受欢迎的十大TED演讲
https://mp.weixin.qq.com/s?__biz=MzIwMTgwNjgyOQ==&mid=2247486529&idx=1&sn=d6136a4558f5152fa2e989b885731838
教程|用人工智能帮你找论文
https://mp.weixin.qq.com/s?__biz=MzIwMTgwNjgyOQ==&mid=2247486619&idx=1&sn=189ef30579aa1810a6396cfc9574be40
20180329:
美团“猜你喜欢”深度学习排序模型实践
https://mp.weixin.qq.com/s?__biz=MjM5NjQ5MTI5OA==&mid=2651747804&idx=1&sn=883777ca375475ff894a446bf41cf7bc
美团外卖骑手背后的AI技术
https://mp.weixin.qq.com/s?__biz=MjM5NjQ5MTI5OA==&mid=2651747804&idx=2&sn=78dc773bf4f34a93a59af4d6bc7840f7
20180328:
keras实现深度学习模型 进行文本分类
https://github.com/lytforgood/TextClassification
20180327:
有哪些比较好的做异常值检测的方法?
https://www.zhihu.com/question/38066650
20180326:
干货丨Kaggle Mercari Price比赛分享,语言不仅是算法和公式而已
https://mp.weixin.qq.com/s?__biz=MzI5ODQxMTk5MQ==&mid=2247484737&idx=1&sn=ea440bc8446c5bf75201b550eb2afccf
美女队长“人机写作巅峰对决”数据竞赛夺冠干货分享
https://mp.weixin.qq.com/s?__biz=MzI5ODQxMTk5MQ==&mid=2247484689&idx=1&sn=faafa4c9649567b091b3a98d465480f8
20180325:
R语言 | 关联规则 (包含可视化实现)
https://blog.csdn.net/gjwang1983/article/details/45015203
基于 Keras 预测航班延误
https://mp.weixin.qq.com/s?__biz=MzI3NTA0MzM1OQ==&mid=2651615418&idx=1&sn=c10fb20de0c7e92965e8bf695b0dfa2f
20180324:
用户画像—打用户行为标签
https://zhuanlan.zhihu.com/p/34804615
比你更了解你,浅谈用户画像
https://zhuanlan.zhihu.com/p/27126412
20180323:
如何对用户进行聚类分析?(微调的回答)
https://www.zhihu.com/question/19982667/answer/347473295
从结构到性能,一文概述XGBoost、Light GBM和CatBoost的同与不同
https://zhuanlan.zhihu.com/p/34698733
20180322:
每天数百亿用户行为数据,美团点评怎么实现秒级转化分析?
https://mp.weixin.qq.com/s?__biz=MjM5NjQ5MTI5OA==&mid=2651747755&idx=1&sn=fe6288c9ce334ad88d89d16fbfe5f045
一文彻底搞懂BP算法:原理推导+数据演示+项目实战(上篇)
https://mp.weixin.qq.com/s/LazepY3DUH3UL_MgV5UeJQ
使用Keras进行深度学习:(二)CNN讲解及实践
https://zhuanlan.zhihu.com/p/34822375
20180321:
K-means算法及文本聚类实践
http://www.cnblogs.com/fengfenggirl/p/k-means.html
学界 | 胶囊网络是如何克服卷积神经网络的这些缺点的?
https://zhuanlan.zhihu.com/p/34619520
使用Keras进行深度学习:(一)Keras 入门
https://zhuanlan.zhihu.com/p/34597052
20180320:
用于数据挖掘的聚类算法有哪些,各有何优势?
https://www.zhihu.com/question/34554321
20180319:
从起源到具体算法,这是一份适合所有人读的深度学习综述论文
https://zhuanlan.zhihu.com/p/34463497
首发皇包车(HI GUIDES)精品旅行服务成单预测竞赛(代码)
https://github.com/zlxy9892/DC_hbc
20180318:
CCF O2O优惠券使用预测第一名资料及代码
https://github.com/wepe/O2O-Coupon-Usage-Forecast
CCF-农产品价格预测线上第5,决赛答辩逆袭为第1名的团队
https://github.com/ziminghuang/futuredata-ccf2016
CCF-农产品价格预测第二名代码
https://github.com/xing89qs/CCF_Product
2016CCF-客户用电异常行为分析
https://github.com/AbnerYang/2016CCF-StateGrid
2016CCF-SouGou-大数据精准营销下的用户画像精准识别
https://github.com/AbnerYang/2016CCF-SouGou
https://github.com/dhdsjy/2016_CCFsougou2
https://github.com/prozhuchen/2016CCF-sougou
https://github.com/coderSkyChen/2016CCF_BDCI_Sougou
联通用户轨迹赛题
https://github.com/xuguanggen/2016CCF-unicom
百度云魅族深度学习应用大赛
https://github.com/ypwhs/baiduyun_deeplearning_competition
天池 商场中精确定位用户所在店铺
https://github.com/hrwhisper/LocationAtMall
20180317:
从起源到具体算法,这是一份适合所有人读的深度学习综述论文
https://zhuanlan.zhihu.com/p/34463497
文本分析相关各种文档提取授权
https://github.com/NLPIR-team/NLPIR/tree/master/License/license%20for%20a%20month
20180316:
互联网产品的盈利模式,你了解多少?
http://www.woshipm.com/pmd/959335.html
首次公开!深度学习在知识图谱构建中的应用
https://mp.weixin.qq.com/s?__biz=MzIzOTU0NTQ0MA==&mid=2247487053&idx=1&sn=8b93080d7e9ed81771d7955ca3e65144
深度学习大讲堂推出免费AI课程
https://zhuanlan.zhihu.com/p/34287459
20180315:
特征工程:机器学习中的特征与标签自然语言化
https://zhuanlan.zhihu.com/p/29394867
同时处理不同类型的问题时,机器学习模型应该想什么
https://zhuanlan.zhihu.com/p/29932095
Mercari Price 比赛分享——语言不仅是算法和公式而已
https://zhuanlan.zhihu.com/p/33987753
20180314:
机器学习模型设计五要素
https://zhuanlan.zhihu.com/p/27783413
DeepFM: A Factorization-Machine based Neural Network for CTR Prediction, IJCAI 2017
https://zhuanlan.zhihu.com/p/27999355
20180313:
天涯明月笙(mtianyan)的github,介绍Django,flask等网站搭建,爬虫等项目
https://github.com/mtianyan
Python 3 & Keras 实现基于神经网络的交通流预测
https://www.jianshu.com/p/1d1c5adf43c6
20180312:
产品经理面试习题大汇总
https://zhuanlan.zhihu.com/p/23948793
产品经理面试自我介绍?(知乎)
https://www.zhihu.com/question/23033429
20180311:
GBDT、XGBoost、LightGBM 的使用及参数调优
https://www.jianshu.com/p/0fe45d4e9542
这是一份优美的信息图,吴恩达点赞的deeplearning.ai课程总结
https://zhuanlan.zhihu.com/p/34346816
20180310:
简单实用的pandas技巧:如何将内存占用降低90%
https://zhuanlan.zhihu.com/p/34420427
20180309:
深度学习这么火,快看如何使用TensorFlow基础函数构建信号回测!
https://zhuanlan.zhihu.com/p/34233208
春节期间,你错过了这些开源论文 | Paper + Code 清单
https://zhuanlan.zhihu.com/p/34245934
20180308:
论文解读 | 基于神经网络的知识推理
https://zhuanlan.zhihu.com/p/34344467
论文Express | 淘宝广告是怎么优化的?阿里团队实时竞价系统策略
https://zhuanlan.zhihu.com/p/34352008
20180307:
深度学习大讲堂推出免费AI课程
https://zhuanlan.zhihu.com/p/34287459
南京大学周志华教授综述论文:弱监督学习
https://zhuanlan.zhihu.com/p/34270286
20180306:
终于!Keras官方中文版文档正式发布了
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650738741&idx=1&sn=d4ba0d9f640676c11f9a0682ed9816fa
20180305:
吴恩达导师Michael I.Jordan学术演讲:如何有效避开鞍点(视频+PPT)
https://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651657898&idx=1&sn=1d67fa6ed40fa06ad1b35256d8482f22
Kaggle大神带你上榜单Top2%:点击预测大赛纪实(上)
https://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651655512&idx=1&sn=51e6d799c195d60fa99ffe5e5c82d5a6
Kaggle大神带你上榜单Top2%:点击预测大赛纪实(下)
https://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651655781&idx=1&sn=3cf841ca023387f74b36999cdf093cdb
20180304:
神经网络学习与TensorFlow实践
https://zhuanlan.zhihu.com/mBrain
20180303:
搞机器学习要哪些技能(机器学习入门)
https://segmentfault.com/a/1190000005356857
信用卡“坏账”客户分析(kaggle数据)
https://mp.weixin.qq.com/s?__biz=MzI2MjE3OTA1MA==&mid=2247484779&idx=1&sn=4dbd56ce51f07a52ec219cfa12faa482
智慧航空AI大赛-阿里云算法大赛总结 第一赛季总结
http://www.cnblogs.com/TextEditor/p/7267307.html
20180302:
400分钟!零基础入门机器学习教程限时领取!!!(天池)
https://tianchi.aliyun.com/forum/new_articleDetail.html?spm=5176.100065.5490641.3.31fe67b78Xsib5&postsId=1515
基于变量KS最大化的分箱方案
https://zhuanlan.zhihu.com/p/33742183
20180301:
历史最全GAN网络及其各种变体整理(附论文及代码实现)
https://zhuanlan.zhihu.com/p/34016536
8个步骤,上手解决90%的 NLP 问题
https://zhuanlan.zhihu.com/p/34120684
学习了!谷歌今日上线基于TensorFlow的机器学习速成课程(中文版)
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650738445&idx=1&sn=b2dbe3aa45c253e37b4a81a0ba3dc4a0
20180228:
聊聊机器学习中的那些树
https://zhuanlan.zhihu.com/p/33987488
20180227:
有哪些牛逼的简历模板?
https://www.zhihu.com/question/26265144
把深度学习作为自己的硕士课题,有什么可以做的?
https://www.zhihu.com/question/27393375
20180226:
同时处理不同类型的问题时,机器学习模型应该想什么
https://zhuanlan.zhihu.com/p/29932095
20180225:
用户增长怎么做 - 我在Uber和摩拜的实战经验
https://zhuanlan.zhihu.com/p/33905485
20180224:
数据科学家效率提升必备技巧之Jupyter Notebook篇
https://zhuanlan.zhihu.com/p/33696082
SOHU图文匹配竞赛-方案分享
https://zhuanlan.zhihu.com/p/27425379
20180223:
基于XGBoost的量化金融实战【系列53】
https://zhuanlan.zhihu.com/p/33948430
20180222:
Python XGBoost算法代码实现和筛选特征应用
https://zhuanlan.zhihu.com/p/33931960
20180221:
一些关于kaggle的比赛方法
https://zhuanlan.zhihu.com/p/33903267
20180220:
房屋价格数据采集与分析
http://blog.csdn.net/zbc1090549839/article/details/54692147
TensorFlow和深度学习入门教程(TensorFlow and deep learning without a PhD)
http://blog.csdn.net/xummgg/article/details/69214366
数据科学家需要了解的5大聚类算法
https://mp.weixin.qq.com/s?__biz=MzI0NTE4NjA0OQ==&mid=2658356887&idx=1&sn=8ed54bca7fc19e34c574b56e50c3dbd7
20180219:
机器学习面试干货精讲
https://zhuanlan.zhihu.com/p/32877396
2017年,机器之心贡献过的开源项目
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650738018&idx=1&sn=924c5a619bfe34d04073f7dc407b74db
20180218:
经典 | 深度学习的7大实用技巧
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247493151&idx=1&sn=9b0a03e949f61c42b883695ff62dc07e
深度 | 机器学习中的模型评价、模型选择及算法选择
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247493266&idx=1&sn=d4ca3d3eaba6a1d4c8f1801622f14aa2
20180217:
绝密 | 机器学习老手不会轻易告诉你的12件事儿
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247493088&idx=1&sn=73f0808c11e98158010faa7348eb20aa
如何快速理解GAN?这里有一篇最直观的解读
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247493117&idx=2&sn=a9118d70b4aa50be81e47b9c32655371
20180216:
15大领域,50篇文章,2018年应当这样学习机器学习
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247492885&idx=1&sn=d41903ad3f45394eefd12d943a4847f6
如何在NLP领域干成第一件事
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247492885&idx=2&sn=d0ba63dc82ceb53d3a7c34113be4ba06
20180215:
机器学习、数据挖掘 如何进阶成为大神?
https://www.zhihu.com/question/37256015/answer/318634686
20180214:
Kaggle大神带你上榜单Top2%:点击预测大赛纪实(上)
http://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651655512&idx=1&sn=51e6d799c195d60fa99ffe5e5c82d5a6
Kaggle大神带你上榜单Top2%:点击预测大赛纪实(下)
http://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651655781&idx=1&sn=3cf841ca023387f74b36999cdf093cdb
20180213:
视频 | 从原理到代码,Hinton的胶囊网络该怎么用?(附代码)
http://mp.weixin.qq.com/s?__biz=MzI5NTIxNTg0OA==&mid=2247489699&idx=3&sn=17ccda7285baa5b7356724cea58cf5c1
20180212:
浪叫兽的竞赛经验总结
https://zhuanlan.zhihu.com/p/33740042
干货 | 请收下这份2018学习清单:150个最好的机器学习,NLP和Python教程
https://mp.weixin.qq.com/s?__biz=MzI5NTIxNTg0OA==&mid=2247489462&idx=4&sn=096557b4a98560528d65a30a47ac037d
20180211:
机器学习顶级论文及实现(附地址及简介)
https://zhuanlan.zhihu.com/p/33720111
YJango:TensorFlow高层API Custom Estimator建立CNN+RNN的演示
https://zhuanlan.zhihu.com/p/33681224
20180210:
如何写好一篇高质量的IEEE/ACM Transaction级别的计算机科学论文?
https://www.zhihu.com/question/22790506/answer/316005175
春节囤货清单 | 15篇近期值得读的AI论文
https://zhuanlan.zhihu.com/p/33729052
20180209:
薅资本主义羊毛,用Google免费GPU
https://mp.weixin.qq.com/s?__biz=MzIzNjc1NzUzMw==&mid=2247493706&idx=1&sn=7474e40d5fe9968cbc79fc35b3c220f4
20180208:
数据城堡社区(包含竞赛思路和竞赛代码)
http://www.dcjingsai.com/common/bbs/topicIndex.html
20180207:
手把手:我的深度学习模型训练好了,然后要做啥?
https://zhuanlan.zhihu.com/p/33665078
201805206:
算法工程师养成实录:学习准备+实习+秋招总结
https://zhuanlan.zhihu.com/p/32215970
非科班野路子的2018秋招算法工程师面经:面试实录+人生经验
https://zhuanlan.zhihu.com/p/31217804
20180205:
机器学习(Machine Learning)&深度学习(Deep Learning)资料(Chapter 1)
https://github.com/ty4z2008/Qix/blob/master/dl.md
数据挖掘竞赛的套路就在这里了,看完本文全明白!
https://mp.weixin.qq.com/s?__biz=MjM5ODU3OTIyOA==&mid=2650669620&idx=2&sn=158af9884aa6ef5d7a4317d3fff2ad09
20180204:
《小王爱迁移》系列之零:迁移学习领域著名学者和研究机构
https://zhuanlan.zhihu.com/p/30685086
20180203:
爬虫进阶教程:百万英雄答题辅助系统
http://cuijiahua.com/blog/2018/01/spider_3.html
Python3网络爬虫(八):爱奇艺等主流视频网站的VIP视频破解(在线观看+视频下载)
http://blog.csdn.net/c406495762/article/details/71334633
Python3网络爬虫实战:VIP视频破解助手;GEETEST验证码破解等
https://github.com/Jack-Cherish/python-spider
爱湃森2017年度Python榜单
https://annual2017.pycourses.com/
20180202:
2017秋招面试总结:计算机视觉/深度学习算法
https://zhuanlan.zhihu.com/p/33020995
大佬面试总结:图像处理/CV/ML/DL到HR面
https://zhuanlan.zhihu.com/p/32981626
《我的PaddlePaddle学习之路》笔记四——自定义图像数据集的识别
http://blog.csdn.net/qq_33200967/article/details/79095265
20180201:
互联网广告综述之点击率特征工程
http://blog.csdn.net/mytestmy/article/details/19088827
2017年度最值得读的AI论文 | NLP篇 (PaperWeekly)
https://zhuanlan.zhihu.com/p/33469791
20180131:
关于sklearn中的决策树是否应该用one-hot编码?(柯国霖等人的回答)
https://www.zhihu.com/question/266195966/answer/306104444
20180130:
500款各领域机器学习数据集,总有一个是你要找的
https://mp.weixin.qq.com/s?__biz=MzI0OTQyNzEzMQ==&mid=2247486065&idx=2&sn=2a4188a9513315278e3681b3fcc1194e
20180129:
微软亚洲研究院论文解读:GAN在网络特征学习中的应用(PPT+视频)
https://zhuanlan.zhihu.com/p/33227420
N问广告ctr预估
https://zhuanlan.zhihu.com/p/33372485
20180128:
Facebook如何运用机器学习进行亿级用户数据处理
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247492743&idx=1&sn=70b557cdf190e76f911cb7e11e86f989
电力AI大赛|机器学习处理流程、特征工程,模型设计实例
https://mp.weixin.qq.com/s?__biz=MzI0OTQyNzEzMQ==&mid=2247486056&idx=1&sn=a8cc5f30ecfb9b2c95874349830f0839
http://www.cnblogs.com/csj007523/p/8310763.html
20180127:
基于Python预测股价的那些人那些坑,请认真看完!【系列52】
https://zhuanlan.zhihu.com/p/33297567
BAT1000题精选 | RNN是怎么从单层网络一步步构造的
https://zhuanlan.zhihu.com/p/33259234
20180126:
就喜欢看综述论文:情感分析中的深度学习
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650736673&idx=1&sn=d3c880d5a11a0865f5a4b0990c306344
深度学习在文本分类中的应用
https://mp.weixin.qq.com/s?__biz=MzI0OTQyNzEzMQ==&mid=2247486054&idx=1&sn=713b0cb002bf0f5ac216203dafc2e5ec
20180125:
爬虫小工具-copyheader
https://zhuanlan.zhihu.com/p/33237247
揭开迷雾,来一顿美味的「Capsule」盛宴 | 附代码实现 + 全新实验
https://zhuanlan.zhihu.com/p/33244896
20180124:
【2018展望Top10】GAN应用落地,NLP急需突破
https://mp.weixin.qq.com/s?__biz=MzI3MTA0MTk1MA==&mid=2652011186&idx=3&sn=63dc532ef61eb2481344d34224124c14
EM算法的九层境界:?Hinton和Jordan理解的EM算法
https://mp.weixin.qq.com/s?__biz=MzI3MTA0MTk1MA==&mid=2652010007&idx=4&sn=8abd041d3b1c5918ed7c4f9a981284e3
【干货】3万字剖析强化学习在电商环境下应用
https://mp.weixin.qq.com/s?__biz=MzI3MTA0MTk1MA==&mid=2652011375&idx=3&sn=6e06c0156de277dee39c02e8420d7111
20180123:
R语言金融波动率建模|基于SGED分布的变参数ARIMA+EARCH动态预测模型的研究
https://mp.weixin.qq.com/s?__biz=MzA3MTM3NTA5Ng==&mid=2651057491&idx=1&sn=ad380aba4b1a78d4e8898e7be75af808
干货 | 如何在NLP领域干成第一件事?
https://mp.weixin.qq.com/s?__biz=MzI5NTIxNTg0OA==&mid=2247488959&idx=2&sn=231637701bb9cad90cddc80fb59c9be2
GANs 造假领衔,看 2018 最具落地潜力的 9 大 AI 技术趋势
https://mp.weixin.qq.com/s?__biz=MzI5NTIxNTg0OA==&mid=2247489179&idx=1&sn=622a79f5d016687c7438c8054666cd5d
学界 | 从ICLR2017被拒论文谈起:行走在GAN的Latent Space
https://mp.weixin.qq.com/s?__biz=MzI5NTIxNTg0OA==&mid=2247489180&idx=2&sn=897bea690a84921d401a3b4ee5ca6983
20180122:
啤酒和尿布的故事是真的吗(关联分析)
https://mp.weixin.qq.com/s?__biz=MzIwNjU2NDcyMg==&mid=2247484084&idx=1&sn=314f3162ae04a7335f9a65929651a3b6
还在用PCA降维?快学学大牛最爱的t-SNE算法吧(附Python/R代码)
https://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651656490&idx=1&sn=2d261af429e87be2240f3c2f133474e6
20180121:
评分卡模型构建介绍
https://zhuanlan.zhihu.com/p/27770760
20180120:
跟繁琐的命令行说拜拜!Gerapy分布式爬虫管理框架来袭!
https://zhuanlan.zhihu.com/p/33117192
20180119:
Python三大评分卡之行为评分卡
https://mp.weixin.qq.com/s?__biz=MzI5NDY1MjQzNA==&mid=2247485355&idx=1&sn=683aa76ab99ed7321ea01c149b83aeb7
20180118:
入门 | 从遗传算法到强化学习,一文介绍五大生物启发式学习算法
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650736276&idx=4&sn=7918365b4b7de6544944189ee43f5fab
学界 | 谷歌大脑提出Adversarial Spheres:从简单流形探讨对抗性样本的来源
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650736232&idx=5&sn=466a31fd21af0fe9fa91b0e1dc6fa3f7
入门 | 想实现DCGAN?从制作一张门票谈起!
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650736232&idx=2&sn=04b1deed56d718decdf271f166d209e2
20180117:
单样本学习(One shot learning)和孪生网络(Siamese Network)简介
https://zhuanlan.zhihu.com/p/29058453
教程 | 如何优雅而高效地使用Matplotlib实现数据可视化
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650736232&idx=4&sn=3332215f55785b316965f1c004d74508
20180116:
机器学习之特征选择
https://www.cnblogs.com/nolonely/p/6435083.html
20180115:
Tensorflow中learning rate decay的奇技淫巧
https://zhuanlan.zhihu.com/p/32923584
AAAI 2018 | 南京大学提出用于聚类的最优间隔分布机
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650735935&idx=4&sn=d4da5646afdb9de7b8891bdcf02e98ce
20180114:
从机器学习谈起
https://mp.weixin.qq.com/s?__biz=MzI2NjA3NTc4Ng==&mid=2652079207&idx=1&sn=04a0703b05d7537a9eb2c8ea3ca457fa
xgboost的原理没你想像的那么难
https://www.jianshu.com/p/7467e616f227
提高驾驶技术:用GAN去除(爱情)动作片中的马赛克和衣服
https://zhuanlan.zhihu.com/p/27199954
20180113:
每个Kaggle冠军的获胜法门:揭秘Python中的模型集成
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650736180&idx=1&sn=4e26c6a0b3c72fd82809ce94b476c55a
互联网征信中的信用评分模型——基于用户在线简历数据
https://mp.weixin.qq.com/s?__biz=MzA5MjEyMTYwMg==&mid=2650240783&idx=1&sn=9c7bfe14bdcc87f0ec0eff6bd109ddf8
20180112:
Kaggle大神带你上榜单Top2%:点击预测大赛纪实(下)
https://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=2651655781&idx=1&sn=3cf841ca023387f74b36999cdf093cdb
20180111:
斯坦福完全可解释深度神经网络:你需要用决策树搞点事
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650735975&idx=1&sn=af91c24d8b0d058f35ac8b1ea65a0ea9
一文读懂「Attention is All You Need」| 附代码实现
https://mp.weixin.qq.com/s?__biz=MzIwMTc4ODE0Mw==&mid=2247486960&idx=1&sn=1b4b9d7ec7a9f40fa8a9df6b6f53bbfb
20180110:
把深度学习作为自己的硕士课题,有什么可以做的?
https://www.zhihu.com/question/27393375/answer/36607524
R语言数据清洗实战——高效list解析方案
https://mp.weixin.qq.com/s?__biz=MzA3MTM3NTA5Ng==&mid=2651057415&idx=1&sn=9c1c7deacd2606064eda6090122a425e
20180109:
动手写机器学习算法:K-Means聚类算法
https://zhuanlan.zhihu.com/p/32776800
机器学习算法实践-Platt SMO和遗传算法优化SVM
https://zhuanlan.zhihu.com/p/30173372
爱奇艺个性化推荐排序实践
https://mp.weixin.qq.com/s?__biz=MzI0MjczMjM2NA==&mid=2247483872&idx=1&sn=db0fbb2bec0d4e68593f1b9bfc20a8b5
20180108:
信用评分模型中的滚动率分析
https://mp.weixin.qq.com/s?__biz=MzI2MDA3NjQwOA==&mid=2653212697&idx=1&sn=05c9a2d890c47e16ee878c89bac8678e
20180107:
用简单的图像识别来实现自动玩微信 “跳一跳” 小游戏
https://zhuanlan.zhihu.com/p/32658168
什么是迁移学习?它都用在深度学习的哪些场景上?这篇文章替你讲清楚了
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247491908&idx=2&sn=e2b0864f07ce56e36e6309972f7b738e
使用word2vec对新浪微博进行情感分析和分类
http://blog.csdn.net/liugallup/article/details/51164962
20180106:
Python网络爬虫实战项目大全!
https://zhuanlan.zhihu.com/p/32546216
20180105:
【干货】--手把手教你完成文本情感分类
https://mp.weixin.qq.com/s?__biz=MzA3MTM3NTA5Ng==&mid=2651057369&idx=2&sn=a8f152d88cbf821edb5de0ae348f6b6d
20180104:
【精心解读】关于Jupyter Notebook的28个技巧
https://zhuanlan.zhihu.com/p/32600329
【精选干货】2017年12月份机器学习排名前10名文章(论文+代码)
https://zhuanlan.zhihu.com/p/32600161
【解读】2017年ML/NLP论文发表情况
https://zhuanlan.zhihu.com/p/32599773
20180103:
自我发掘爬虫实战1:宅男女神网妹子图片批量抓取,分类保存到本地和MongoDB数据库
https://zhuanlan.zhihu.com/p/31421316
盘点深度学习一年来在文本、语音和视觉等方向的进展,看强化学习如何无往而不利
https://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247491981&idx=1&sn=c9d6f9837722103d9f513b0bb8ec282e
行业应用 | 自然语言理解如何解放金融从业者
https://zhuanlan.zhihu.com/p/32201875
20180102:
评分卡模型构建介绍
https://zhuanlan.zhihu.com/p/27770760
数据切割(分箱,离散化)及woe、iv值计算的python代码分享
https://zhuanlan.zhihu.com/p/32265553
20180101:
教程 | 理解XGBoost机器学习模型的决策过程
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650734959&idx=2&sn=93882c4f286880d7db364ba7b33dd6ce
资源 | XGBoost 中文文档开放:上去就是一把梭
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650735126&idx=4&sn=87981d70be4a864483c51e0d535cce1e
教程 | 如何用Docker成为更高效的数据科学家?
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650735018&idx=4&sn=82759b19c393f46ff780ce08b5000088
前沿 | 利用遗传算法优化神经网络:Uber提出深度学习训练新方式
https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2650735095&idx=4&sn=5d33ab2a083e49ad63fd73a1e982a650
#2018年的内容浏览完毕,如果想继续查找数据挖掘,人工智能,机器学习,深度学习方面的资料可以前往2017Record_1.md和2017Record_2.md