基于向量的召回算法及其在个性化广告新闻中的应用实践-刘政
31页1、 Embedding Based Recall: Practice,Progress and PerspectivesZheng Liu , Jianxun Lian, Xing XieSocial Computing Group, MSRAAug 15 , 2021thReinforced Anchor Knowledge Graph Generation for News Recommendation Reasoning, Liu et. al.KDD 2021 Outline Overview Multi-Stage Pipeline EBR: Pros and Cons Embedding learning algorithms Negative Augmentation Hard Negative Sampling Diversified representation Training as knowledge distillation Things beyond learning algorithms Efficiency issues Combo of sparse an
2、d dense Overview: Multi-Stage PipelineL1 Stage(Recall)L2 Stage(Rank)L3 Stage(Re-rank) 1001 1069 1013 Rank : high-precision, KPI-oriented Recall : fast, accurate, comprehensive Overview: Multi-Stage PipelineFast, AccurateAdsQueryXbox 360 4GB SlimConsoleMicrosoft Xbox 360Microsoft Xbox 360 E250GBXbox 360 Game SystemHDMI Overview: Multi-Stage PipelineVocabulary mis-matchAdsQueryXbox 360 4GB SlimConsoleMicrosoft gameMicrosoft Xbox 360 Econsole250GBXbox 360 Game SystemHDMI Overview: EBR, Pros and Cos
3、High-generalizable,Relatively fastMicrosoft gameconsoleXbox 360 Game SystemANN Index (PQ, HNSW)HDMI Overview: EBR, Pros and ConsModels training isdata-intensive Overview : EBR, Pros and Consembeddings canbeambiguousNintendo switchconsole the Xbox 360 console harddrive Xbox 360 GameSystem HDMIHard drive will be at least 20GBmodel supports HD graphics in 16 x 9wide-screen, with anti-aliasing Overview : EBR, Pros and ConsAlignmentUniformity1 SimCSE, Gao et.al.2 Understanding ContrastiveLearning ICM
4、L 2020, Wang et. al. Outline Overview Multi-Stage Pipeline EBR: Pros and Cons Embedding learning algorithms Negative Augmentation Hard Negative Sampling Training as distillation Diversified representation Things beyond learning algorithms Efficiency issues Combo of sparse and dense Algos: Negative AugmentationQKOnes positiveusedLarger batch size-others negativehigher accuracyDPR, Karpukhin et. al.batch Algos: Negative AugmentationExpand#negativesby +Dev-1Dev-Cross-device negativesamplingDev-Rock
《基于向量的召回算法及其在个性化广告新闻中的应用实践-刘政》由会员郭**分享,可在线阅读,更多相关《基于向量的召回算法及其在个性化广告新闻中的应用实践-刘政》请在金锄头文库上搜索。
2023汽车行业人才管理与薪酬趋势
2023麦肯锡金融业白皮书:后疫情时代财富管理重启增长
国家金融与发展实验室-数字经济背景下科技服务商对商业银行的异质性影响研究-12页
2023年3月银行在长期背景下的干预——硅谷银行及其他银行(英)-2023.3-10页
2023视频号商业生态发展报告:多元增长迎接爆发
2023新能源汽车行业充电桩行业深度报告:千亿市场盈利边际向好
2023新能源汽车充电桩行业深度报告:千亿市场盈利边际向好
2022乡村旅游振兴白皮:乡村旅游新阶段乡村民宿挑起增长大梁-携程
黑马工厂《坚果行业趋势报告》-33页
海外发展研究所-埃塞俄比亚和AfCFTA的投资条款:问题和挑战(英)-2022.12-36页
推动药物开发的创新安进公司揭示三大研发方向-117页
《追求成功的热情》
营销数字化转型登山图(2021版)-66页
《稻盛和夫为何能不断成功》
未来远程混合工作模式的挑战和机会(2021)-50页
商用密码应用安全性评估白皮书-69页
世界主要能源数据统计2021-81页
2022服装行业出海营销报告-71页
2022爱分析·人工智能厂商全景报告-85页
网络数据安全管理条例-208
2023-07-21 51页
2023-07-21 61页
2023-07-21 396页
2023-05-17 12页
2023-05-17 10页
2023-05-05 207页
2023-05-05 69页
2023-05-05 70页
2023-03-09 34页
2023-03-09 43页