大数据哈希学习南京大学李武军博士的PPT

【实例简介】
大数据哈希学习南京大学李武军博士的PPT，信息量非常大，哈希学习的资料少，非常难得，推荐下载！！
Outline Introduction Unsupervised Hashing Supervised Hashing Ranking-based Hashing Multimodal Hashing Deep hashi Quantization Reference EQC Li(http://cs.nju.edu.cn/lwj) Learning to Hash S. NJU 3/210 Nearest Neighbor Search(Retrieval) o Given a query point q, return the points closest(similar to q in the database(e.g, image database) o Underlying many machine learning, data mining, information retrieva roble EQC Li(http://cs.nju.edu.cn/lwj) Learning to Hash CS. NJU 4/210 Big data Big data has attracted much attention from both academia and industry o Facebook: 750 million users o Flickr: 6 billion photos o Wal-Mart: 267 million items day 4Pb data warehouse o Sloan Digital Sky Survey: New mexico telescope captures 200 GB Image data/day nature Science FOURTH PARADIGM SCIENCEIN TH PETABYTEERA data EQC Li(http://cs.nju.educn/lwj Learning to Hash S. NJU 5/210 Nearest Neighbor Search(NNS) for Big Data Challenge in big data applications o Curse of dimensionality ● Storage cost Query speec EQC Li(http://cs.nju.edu.cn/lwj) Learning to Hash S. NJU 6/210 Similarity Preserving Hashing h(Statue of liberty) h(Napoleon) h(napoleon) 10001010 01100001 01100101 Should be very different Should be similar Li(http://cs.nju.educn/lwj Learning to Hash CS. NJU 7/210 Reduce dimensionality and storage cost Bil 10 million image 20 GB 160MB 512valu EQC Li(http://cs.nju.edu.cn/lwj) Learning to Hash S. NJU 8/210 Fast Query Speed o By using hash-code to construct index, we can achieve constant or sub-linear search time complexity o In some cases, exhaustive search with linear time complexity is also acceptable because the distance calculation cost is low with binary representation EQC Li(http://cs.nju.educn/lwj Learning to Hash CS. NJU 9/210 Two Stages of Hash Function Learning Two main categories Category I Projection Stage(Dimension Reduction rejected with real-valued projection function o Given a point x, each projected dimension i will be associated with a tion function∫(x)(e.g…,f2(x)=wx) Quantization stage o Essential difference between metric learning and learning to hash Category Il Binary-Code Learning Stage Hash Function Learning Stage EQC Li(http://cs.nju.educn/lwj Learning to Hash CS. NJU 10/210 【实例截图】
【核心代码】