《论文阅读笔记》Deep Metric Learning via Lifted Structured Feature Embedding

1.解释Lifted ,Structured

structured:结构化
可以参考structured learning : to prediction structured ojects ,rather than scalar discrete or real values.For example,the problem of translating a natural language sentence into a syntactic representation such as a parse tree can be seen as a structured prediction problem in which the structured output domain is the set of all possible parse trees.

feature embedding:特征映射

Lifting problem : 参考 Homotopy lifting property
wiki:https://en.wikipedia.org/wiki/Homotopy_lifting_property

2.what is metric learning
度量学习,也就是为了学习一个度量函数,也是一个相似性函数,使得相似的物体的语义距离近,不相似的物体的语义距离远.
当前主流的Metric learning 方法当属triplet loss,常见的还有Contrastive loss,但是这两种方法都没有很充分的利用每个training batch 的信息。引用原文的一句话:
taking full advantage of the training batches in the neural network training by lifting the vector of pairwise distance within the batch to the matrix of pairwise distances

3.what is the problem
the existing approaches cannot take full adwantage of the training batches used during the mini-batch stochastic gradient descent traning of the networks 。即,由于采用的是deep learning 的方法,我们采用的SGD随机梯度方法训练网络,每次计算都是一个batch,triplet 和contrastive 这这两种方法都没有充分利用整个batch size,triplet 是一个三元组(achor,negative,postive),contrastive是一个二元组(xi,xj)(xi,xj可以是来自同一个类,或者来自不同类),如果是triplet 和 contrastive ,每个batch size 都是由这些二元组或者三元组组成,如下图
这里写图片描述

计算loss 和更新权重时,只依赖(x1,x2),(x3,x4),(x5,x6),而跟(x1,x3),(x4,x6)等等没有关系,triplet也有这种问题。

为了充分利用一个batch 里样本,来更新权重和计算loss,作者就提出了 Lifted Structured embedding

4 . what is Lifted Structured embedding

如果想学习最完美的metric learning 函数,那么理论上的loss 函数:
这里写图片描述

P是train set里的正样本对集,N是train set里所有可能的负样本对集,但是如果想用deep learning来实现的话,那么可以看到,负样本对太多了。
所以这样就导致了两个问题:
(1)loss 是一个非光滑函数
(2)计算或者估计次梯度需要频繁的求最小的pairs,而且是再整个样本集。

那么怎么解决这个问题呢:
(1)将loss 换成一个光滑的上界函数来代替;
(2)将样本集分成无数小的batch,每个batch 采用上述方法计算loss

因此从之前的O(m)pairs变成了O(m^2) 个pairs

5 .定义距离函数

这里写图片描述

f(x)是cnn 学习到的特征,将每张图片映射到这个256维的向量,然后用这256维向量来计算距离。

6.光滑的上界函数是

这里写图片描述

转化为光滑的上界函数也是因为:嵌套的max loss函数会导致network 收敛到一个坏的局部最优点。

然后就是对着loss函数求导了:

这里写图片描述
这里写图片描述

lifted structured feature embedding 大大增加了训练的样本数目pairs num;

  • 3
    点赞
  • 2
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
Part I Metric Searching in a Nutshell Overview 3 1. FOUNDATIONS OF METRIC SPACE SEARCHING 5 1 The Distance Searching Problem 6 2 The Metric Space 8 3 Distance Measures 9 3.1 Minkowski Distances 10 3.2 Quadratic Form Distance 11 3.3 Edit Distance 12 3.4 Tree Edit Distance 13 3.5 Jaccard’s Coefficient 13 3.6 Hausdorff Distance 14 3.7 Time Complexity 14 4 Similarity Queries 15 4.1 Range Query 15 4.2 Nearest Neighbor Query 16 4.3 Reverse Nearest Neighbor Query 17 4.4 Similarity Join 17 4.5 Combinations of Queries 18 4.6 Complex Similarity Queries 18 5 Basic Partitioning Principles 20 5.1 Ball Partitioning 20 5.2 Generalized Hyperplane Partitioning 21 5.3 Excluded Middle Partitioning 21 5.4 Extensions 21 6 Principles of Similarity Query Execution 22 6.1 Basic Strategies 22 6.2 Incremental Similarity Search 25 7 Policies for Avoiding Distance Computations 26 7.1 Explanatory Example 27 7.2 Object-Pivot Distance Constraint 28 7.3 Range-Pivot Distance Constraint 30 7.4 Pivot-Pivot Distance Constraint 31 7.5 Double-Pivot Distance Constraint 33 7.6 Pivot Filtering 34 8 Metric Space Transformations 35 8.1 Metric Hierarchies 36 8.1.1 Lower-Bounding Functions 36 8.2 User-Defined Metric Functions 38 8.2.1 Searching Using Lower-Bounding Functions 38 8.3 Embedding Metric Space 39 8.3.1 Embedding Examples 39 8.3.2 Reducing Dimensionality 40 9 Approximate Similarity Search 41 9.1 Principles 41 9.2 Generic Algorithms 44 9.3 Measures of Performance 46 9.3.1 Improvement in Efficiency 46 9.3.2 Precision and Recall 46 9.3.3 Relative Error on Distances 48 9.3.4 Position Error 49 10 Advanced Issues 50 10.1 Statistics on Metric Datasets 51 10.1.1 Distribution and Density Functions 51 10.1.2 Distance Distribution and Density 52 10.1.3 Homogeneity of Viewpoints 54 10.2 Proximity of Ball Regions 55 10.3 Performance Prediction 58 Contents ix 10.4 Tree Quality Measures 60 10.5 Choosing Reference Points 63 2. SURVEY OF EXISTING APPROACHES 67 1 Ball Partitioning Methods 67 1.1 Burkhard-Keller Tree 68 1.2 Fixed Queries Tree 69 1.3 Fixed Queries Array 70 1.4 Vantage Point Tree 72 1.4.1 Multi-Way Vantage Point Tree 74 1.5 Excluded Middle Vantage Point Forest 75 2 Generalized Hyperplane Partitioning Approaches 76 2.1 Bisector Tree 76 2.2 Generalized Hyperplane Tree 77 3 Exploiting Pre-Computed Distances 78 3.1 AESA 78 3.2 Linear AESA 79 3.3 Other Methods 80 4 Hybrid Indexing Approaches 81 4.1 Multi Vantage Point Tree 81 4.2 Geometric Near-neighbor Access Tree 82 4.3 Spatial Approximation Tree 85 4.4 M-tree 87 4.5 Similarity Hashing 88 5 Approximate Similarity Search 89 5.1 Exploiting Space Transformations 89 5.2 Approximate Nearest Neighbors with BBD Trees 90 5.3 Angle Property Technique 92 5.4 Clustering for Indexing 94 5.5 Vector Quantization Index 95 5.6 Buoy Indexing 97 5.7 Hierarchical Decomposition of Metric Spaces 97 5.7.1 Relative Error Approximation 98 5.7.2 Good Fraction Approximation 98 5.7.3 Small Chance Improvement Approximation 98 5.7.4 Proximity-Based Approximation 99 5.7.5 PAC Nearest Neighbor Search 99 x SIMILARITY SEARCH Part II Metric Searching in Large Collections of Data Overview 103 3. CENTRALIZED INDEX STRUCTURES 105 1 M-tree Family 105 1.1 The M-tree 105 1.2 Bulk-Loading Algorithm of M-tree 109 1.3 Multi-Way Insertion Algorithm 112 1.4 The Slim Tree 113 1.4.1 Slim-Down Algorithm 114 1.4.2 Generalized Slim-Down Algorithm 116 1.5 Pivoting M-tree 118 1.6 The M

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值