SlideShare ist ein Scribd-Unternehmen logo
1 von 23
1
Hashing: Object Embedding
Reporter: Xu Jiaming (PH.D Student)
Date: 2014.03.27
Computational-Brain Research Center
Institute of Automation, Chinese Academy of Sciences
Report
2
First, What is Embedding?
[ ć‡șè‡Ș ]: https://en.wikipedia.org/wiki/Embedding
When some object X is said to be embedded in another object Y,
the embedding is given by some injective and structure-
preserving map f : X → Y. The precise meaning of "structure-
preserving" depends on the kind of mathematical structure of which
X and Y are instances.
Structure-Preserving in IR:
1 2 1 2
:
(X ,X ) (Y ,Y )
f
Sim Sim
→
≈
X Y
3
Then, What is Hash?
[ ć‡șè‡Ș ]: https://en.wikipedia.org/wiki/Hash_table
The hash function will assign each key to a unique bucket, but this
situation is rarely achievable in practice (usually some keys will
hash to the same bucket). Instead, most hash table designs assume
that hash collisions—different keys that are assigned by the hash
function to the same bucket—will occur and must be
accommodated in some way.
4
Combine the Two Properties
[1998, Piotr Indyk, cited: 1847]
Locality Sensitive Hashing
. ( , ) , . Pr[ ( ) ( )] 1
. ( , ) (1 ), . Pr[ ( ) ( )] 1
if D p q r then h p h q p
if D p q r then h p h q pΔ
≀ = ≄
> + = ≄
5
Overview of Hashing
Real World Binary Space
2000 values 32 bits
Binary
Reduction
6
Facing Big Data
Approximation
7
Learning to Hash
Data-Oblivious
Data-Aware
Description Methods
LSI, RBM, SpH, STH, 

LSH, Kernel-LSH, SimHash, 

8
Data-Oblivious: SimHash [WWW.2007]
Text




Observed
Features
W1
W2
Wn
100110 W1
110000 W2
001001 Wn




W1 –W1 -W1 W1 W1 -W1
W2 W2 -W2 -W2 -W2 -W2
-Wn –Wn Wn –Wn –Wn Wn



13, 108, -22, -5, -32, 551, 1, 0, 0, 0,
1
Step1:
Compute TF-
IDF
Step2: Hash
Function
Step3: Signature
Step4: Sum
Step5: Generate
Fingerprint
9
Data-Aware: Spectral Hashing [NIPS.2008]
2
min :
. .: { 1,1}
0
1
ij i j
ij
k
i
i
i
T
i i
i
S y y
s t y
y
y y
n
−
∈ −
=
=
∑
∑
∑ I
min : ( ( ) )
. .: ( , ) { 1,1}
0
T
k
T
T
trace Y D W Y
s t Y i j
−
∈ −
=
=
Y 1
Y Y I
Laplacian Eigenmap
XW Y=
10
Some Questions?
1. Can we obtain hashing codes by binarizing the real-valued low-
dimensional vectors such as LSI?
2. Can we get hashing codes by Deep Learning approaches such
as RBM, or AutoEncoder?
11
Some Questions?
1. Can we obtain hashing codes by binarizing the real-valued low-
dimensional vectors such as LSI?
Of Course !
[R. Salakhutdinov, G. Hinton. Semantic Hashing, SIGIR2007]
2. Can we get hashing codes by Deep Learning approaches such
as RBM, or AutoEncoder?
No Problem !
[R. Salakhutdinov, G. Hinton. Semantic Hashing, SIGIR2007]
12
In 2013, What Did They Think About?
Total: 30
13
1/9 - ICML2013:
Title: Learning Hash Functions Using Column Generation
Authors: Xi Li, Guosheng Lin, Chunhua Shen, Anton van den Hengel, Anthony Dick
Organization: The University of Adelaide (Australia)
Based On: NIPS2005: Distance Metric Learning for Large Margin Nearest Neighbor ClassiïŹcation
Motivation: In content based image retrieval, to collect feedback, users may be required to report
whether image x looks more similar to x+ than it is to a third image x−. This task is typically much
easier than to label each individual image.
11
min
. . 0, 0;
( , ) ( , ) 1 ,
J
i i
H i i H i i i
C
s t
d d i
Ο
Ο
Ο
=
− +
+
≄ ≄
− ≄ − ∀
∑w,Ο
w
w
x x x x
14
2/9 - ICML2013:
Title: Predictable Dual-View Hashing
Authors: Mohammad Rastegari, Jonghyun Choi, Shobeir Fakhraei, Hal Daume III, Larry S. Davis
Organization: The University of Maryland (USA)
Motivation: It is often the case that information about data are available from two or more views, e.g.,
images and their textual descriptions. It is highly desirable to embed information from both domains in
the binary codes, to increase search and retrieval capabilities.
2 2 2 2
2 2 2 2
min
. . sgn(W X )
sgn(W X )
T T T T
V V T T T T T V V V
T
T T T
T
V V V
W X Y Y Y I W X Y Y Y I
s t Y
Y
− + − + − + −
=
=
Y,U
15
3/9 - SIGIR2013:
Title: Semantic Hashing Using Tags and Topic Modeling.
Authors: Qifan Wang, Dan Zhang, Luo Si
Organization: Purdue University (USA)
Motivation: Two major issues are not addressed in the existing hashing methods: (1) Tag information
is not fully utilized in previous methods. Most existing methods only deal with the contents of
documents without utilizing the information contained in tags; (2) Document similarity in the
original keyword feature space is used as guidance for generating hashing codes in previous methods,
which may not fully reflect the semantic relationship.
1
2
2 2 2
min ( )
. . { 1,1} , 0
T
F
k n
C
s t
Îł
×
− + + −
∈ − =
Y,U
T U Y U YΞ
Y Y1
g
16
3/9 - SIGIR2013:
Title: Semantic Hashing Using Tags and Topic Modeling.
Authors: Qifan Wang, Dan Zhang, Luo Si
Organization: Purdue University (USA)
Motivation: Two major issues are not addressed in the existing hashing methods: (1) Tag information
is not fully utilized in previous methods. Most existing methods only deal with the contents of
documents without utilizing the information contained in tags; (2) Document similarity in the
original keyword feature space is used as guidance for generating hashing codes in previous methods,
which may not fully reflect the semantic relationship.
1
2
2 2 2
min ( )
. . { 1,1} , 0
T
F
k n
C
s t
Îł
×
− + + −
∈ − =
Y,U
T U Y U YΞ
Y Y1
g
Our experiments on
20Newsgroups
17
4/9 - IJCAI2013:
Title: A Unified Approximate Nearest Neighbor Search Scheme by Combining Data Structure and
Hashing.
Authors: Debing Zhang, Genmao Yang, Yao Hu, Zhongming Jin, Deng Cai, Xiaofei He
Organization: Zhejiang University (China)
Motivation: Traditionally, to solve problem of nearest neighbor search, researchers mainly focus on
building effective data structures such as hierarchical k-means tree or using hashing methods to
accelerate the query process. In this paper, we propose a novel unified approximate nearest neighbor
search scheme to combine the advantages of both the effective data structure and the fast Hamming
distance computation in hashing methods.
18
5/9 - CVPR2013:
Title: K-means Hashing: an Affinity-Preserving Quantization Method for Learning Binary
Compact Codes.
Authors: Kaiming He, Fang Wen, Jian Sun
Organization: Microsoft Research Asia (China)
Motivation: Both Hamming-based methods and lookup-based methods are of growing interest
recently, and each category has its benefits depending on the scenarios. The lookup-based methods
have been shown more accurate than some Hamming methods with the same code-length. However,
the lookup-based distance computation is slower than the Hamming distance computation. Hamming
methods also have the advantage that the distance computation is problem-independent
1 1
2
0 0
( ( , ) ( , ))
k k
aff ij i j h
i j
E w d c c d i j
− −
= =
= −∑∑
19
6/9 - ICCV2013:
Title: Complementary Projection Hashing.
Authors: Zhongming Jin1
, Yao Hu1
, Yue Lin1
, Debing Zhang1
, Shiding Lin2
, Deng Cai1
, Xuelong Li3
Organization: 1. Zhejiang University, 2. Baidu Inc., 3. Chinese Academy of Sciences, Xi’an (China)
Motivation: 1. (a) The hyperplane a crosses the sparse region and the neighbors are quantized into the
same bucket; (b) The hyperplane b crosses the dense region and the neighbors are quantized into the
different buckets. Apparently, the hyperplane a is more suitable as a hashing function. 2. (a) (b) Both
the hyperplane a and the hyperplane b can evenly separated the data. (c) However, putting them
together does not generate a good two bits hash function. (d) A better example for two bits hash
function
20
7/9 - CVPR2013:
Title: Hash Bit Selection: a Unified Solution for Selection Problems in Hashing.
Authors: Xianglong Liu1
, Junfeng He2,3
, Bo Lang1
, Shih-Fu Chang2
.
Organization: 1. Beihang University(China), 2. Columbia University(US), 3. Facebook(US)
Motivation: Recent years have witnessed the active development of hashing techniques for nearest
neighbor search over big datasets. However, to apply hashing techniques successfully, there are
several important issues remaining open in selection features, hashing algorithms, parameter
settings  kernels, etc.
21
8/9 - ICCV2013:
Title: A General Two-Step Approach to Learning-Based Hashing.
Authors: Guosheng Lin, Chunhua Shen, David Suter, Anton van den Hengel.
Organization: University of Adelaide (Australia)
Based On: SIGIR2010: Self-Taught Hashing for Fast Similarity Search
Motivation: Most existing approaches to hashing apply a single form of hash function, and an
optimization process which is typically deeply coupled to this specific form. This tight coupling
restricts the flexibility of the method to respond to the data, and can result in complex optimization
problems that are difficult to solve. Their framework decomposes hashing learning problem into two
steps: hash bit learning and hash function learning based on the learned bits.
22
9/9 - IJCAI2013:
Title: Smart Hashing Update for Fast Response.
Authors: Qiang Yang, Long-Kai Huang, Wei-Shi Zheng, Yingbiao Ling.
Organization: Sun Yat-sen University (China)
Based On: DMKD2012: Active Hashing and Its Application to Image and Text Retrieval
Motivation: Although most existing hashing-based methods have been proven to obtain high accuracy,
they are regarded as passive hashing and assume that the labeled points are provided in advance. In this
paper, they consider updating a hashing model upon gradually increased labeled data in a fast response
to users, called smart hashing update (SHU).
1. Consistency-based Selection;
2. Similarity-based Selection.
[CVPR.2012]
( , ) min{ ( , , 1), ( , ,1)}Diff k j num k j num k j= −
2
{ 1,1}
1
min l r
l
T
l l
H
F
Q H H S
r×
∈ −
= −
2
1 1
{1,2,...,r}
min k k T
k r r Fk
R rS H H− −
∈
= −
23
Reporter: Xu Jiaming (Ph.D Student)
Date: 2014.03.27
Computational-Brain Research Center
Institute of Automation, Chinese Academy of Sciences

Weitere Àhnliche Inhalte

Was ist angesagt?

Lifelong Topic Modelling presentation
Lifelong Topic Modelling presentation Lifelong Topic Modelling presentation
Lifelong Topic Modelling presentation Daniele Di Mitri
 
Basic review on topic modeling
Basic review on  topic modelingBasic review on  topic modeling
Basic review on topic modelingHiroyuki Kuromiya
 
Fusing semantic data
Fusing semantic dataFusing semantic data
Fusing semantic dataAndriy Nikolov
 
Distribution Similarity based Data Partition and Nearest Neighbor Search on U...
Distribution Similarity based Data Partition and Nearest Neighbor Search on U...Distribution Similarity based Data Partition and Nearest Neighbor Search on U...
Distribution Similarity based Data Partition and Nearest Neighbor Search on U...Editor IJMTER
 
Hierarchal clustering and similarity measures along with multi representation
Hierarchal clustering and similarity measures along with multi representationHierarchal clustering and similarity measures along with multi representation
Hierarchal clustering and similarity measures along with multi representationeSAT Journals
 
Hierarchal clustering and similarity measures along
Hierarchal clustering and similarity measures alongHierarchal clustering and similarity measures along
Hierarchal clustering and similarity measures alongeSAT Publishing House
 
Mining Users Rare Sequential Topic Patterns from Tweets based on Topic Extrac...
Mining Users Rare Sequential Topic Patterns from Tweets based on Topic Extrac...Mining Users Rare Sequential Topic Patterns from Tweets based on Topic Extrac...
Mining Users Rare Sequential Topic Patterns from Tweets based on Topic Extrac...IRJET Journal
 
IRJET- Text Document Clustering using K-Means Algorithm
IRJET-  	  Text Document Clustering using K-Means Algorithm IRJET-  	  Text Document Clustering using K-Means Algorithm
IRJET- Text Document Clustering using K-Means Algorithm IRJET Journal
 
Learning possibilistic networks from data: a survey
Learning possibilistic networks from data: a surveyLearning possibilistic networks from data: a survey
Learning possibilistic networks from data: a surveyUniversity of Nantes
 
Concurrent Inference of Topic Models and Distributed Vector Representations
Concurrent Inference of Topic Models and Distributed Vector RepresentationsConcurrent Inference of Topic Models and Distributed Vector Representations
Concurrent Inference of Topic Models and Distributed Vector RepresentationsParang Saraf
 
Slides: Concurrent Inference of Topic Models and Distributed Vector Represent...
Slides: Concurrent Inference of Topic Models and Distributed Vector Represent...Slides: Concurrent Inference of Topic Models and Distributed Vector Represent...
Slides: Concurrent Inference of Topic Models and Distributed Vector Represent...Parang Saraf
 
An Improved Similarity Matching based Clustering Framework for Short and Sent...
An Improved Similarity Matching based Clustering Framework for Short and Sent...An Improved Similarity Matching based Clustering Framework for Short and Sent...
An Improved Similarity Matching based Clustering Framework for Short and Sent...IJECEIAES
 
EXPERT OPINION AND COHERENCE BASED TOPIC MODELING
EXPERT OPINION AND COHERENCE BASED TOPIC MODELINGEXPERT OPINION AND COHERENCE BASED TOPIC MODELING
EXPERT OPINION AND COHERENCE BASED TOPIC MODELINGijnlc
 
Simplicial closure & higher-order link prediction
Simplicial closure & higher-order link predictionSimplicial closure & higher-order link prediction
Simplicial closure & higher-order link predictionAustin Benson
 
FAST FUZZY FEATURE CLUSTERING FOR TEXT CLASSIFICATION
FAST FUZZY FEATURE CLUSTERING FOR TEXT CLASSIFICATION FAST FUZZY FEATURE CLUSTERING FOR TEXT CLASSIFICATION
FAST FUZZY FEATURE CLUSTERING FOR TEXT CLASSIFICATION cscpconf
 
Reduct generation for the incremental data using rough set theory
Reduct generation for the incremental data using rough set theoryReduct generation for the incremental data using rough set theory
Reduct generation for the incremental data using rough set theorycsandit
 
Bi4101343346
Bi4101343346Bi4101343346
Bi4101343346IJERA Editor
 
Expandable bayesian
Expandable bayesianExpandable bayesian
Expandable bayesianAhmad Amri
 

Was ist angesagt? (18)

Lifelong Topic Modelling presentation
Lifelong Topic Modelling presentation Lifelong Topic Modelling presentation
Lifelong Topic Modelling presentation
 
Basic review on topic modeling
Basic review on  topic modelingBasic review on  topic modeling
Basic review on topic modeling
 
Fusing semantic data
Fusing semantic dataFusing semantic data
Fusing semantic data
 
Distribution Similarity based Data Partition and Nearest Neighbor Search on U...
Distribution Similarity based Data Partition and Nearest Neighbor Search on U...Distribution Similarity based Data Partition and Nearest Neighbor Search on U...
Distribution Similarity based Data Partition and Nearest Neighbor Search on U...
 
Hierarchal clustering and similarity measures along with multi representation
Hierarchal clustering and similarity measures along with multi representationHierarchal clustering and similarity measures along with multi representation
Hierarchal clustering and similarity measures along with multi representation
 
Hierarchal clustering and similarity measures along
Hierarchal clustering and similarity measures alongHierarchal clustering and similarity measures along
Hierarchal clustering and similarity measures along
 
Mining Users Rare Sequential Topic Patterns from Tweets based on Topic Extrac...
Mining Users Rare Sequential Topic Patterns from Tweets based on Topic Extrac...Mining Users Rare Sequential Topic Patterns from Tweets based on Topic Extrac...
Mining Users Rare Sequential Topic Patterns from Tweets based on Topic Extrac...
 
IRJET- Text Document Clustering using K-Means Algorithm
IRJET-  	  Text Document Clustering using K-Means Algorithm IRJET-  	  Text Document Clustering using K-Means Algorithm
IRJET- Text Document Clustering using K-Means Algorithm
 
Learning possibilistic networks from data: a survey
Learning possibilistic networks from data: a surveyLearning possibilistic networks from data: a survey
Learning possibilistic networks from data: a survey
 
Concurrent Inference of Topic Models and Distributed Vector Representations
Concurrent Inference of Topic Models and Distributed Vector RepresentationsConcurrent Inference of Topic Models and Distributed Vector Representations
Concurrent Inference of Topic Models and Distributed Vector Representations
 
Slides: Concurrent Inference of Topic Models and Distributed Vector Represent...
Slides: Concurrent Inference of Topic Models and Distributed Vector Represent...Slides: Concurrent Inference of Topic Models and Distributed Vector Represent...
Slides: Concurrent Inference of Topic Models and Distributed Vector Represent...
 
An Improved Similarity Matching based Clustering Framework for Short and Sent...
An Improved Similarity Matching based Clustering Framework for Short and Sent...An Improved Similarity Matching based Clustering Framework for Short and Sent...
An Improved Similarity Matching based Clustering Framework for Short and Sent...
 
EXPERT OPINION AND COHERENCE BASED TOPIC MODELING
EXPERT OPINION AND COHERENCE BASED TOPIC MODELINGEXPERT OPINION AND COHERENCE BASED TOPIC MODELING
EXPERT OPINION AND COHERENCE BASED TOPIC MODELING
 
Simplicial closure & higher-order link prediction
Simplicial closure & higher-order link predictionSimplicial closure & higher-order link prediction
Simplicial closure & higher-order link prediction
 
FAST FUZZY FEATURE CLUSTERING FOR TEXT CLASSIFICATION
FAST FUZZY FEATURE CLUSTERING FOR TEXT CLASSIFICATION FAST FUZZY FEATURE CLUSTERING FOR TEXT CLASSIFICATION
FAST FUZZY FEATURE CLUSTERING FOR TEXT CLASSIFICATION
 
Reduct generation for the incremental data using rough set theory
Reduct generation for the incremental data using rough set theoryReduct generation for the incremental data using rough set theory
Reduct generation for the incremental data using rough set theory
 
Bi4101343346
Bi4101343346Bi4101343346
Bi4101343346
 
Expandable bayesian
Expandable bayesianExpandable bayesian
Expandable bayesian
 

Andere mochten auch

Search space reduction for holistic ligature recognition in Urdu Nastaliq scr...
Search space reduction for holistic ligature recognition in Urdu Nastaliq scr...Search space reduction for holistic ligature recognition in Urdu Nastaliq scr...
Search space reduction for holistic ligature recognition in Urdu Nastaliq scr...Akram El-Korashy
 
Memory Networks, Neural Turing Machines, and Question Answering
Memory Networks, Neural Turing Machines, and Question AnsweringMemory Networks, Neural Turing Machines, and Question Answering
Memory Networks, Neural Turing Machines, and Question AnsweringAkram El-Korashy
 
PFI Seminar 2012/03/15 ă‚«ăƒŒăƒăƒ«ăšăƒăƒƒă‚·ăƒ„ăźæ©Ÿæą°ć­Šçż’
PFI Seminar 2012/03/15 ă‚«ăƒŒăƒăƒ«ăšăƒăƒƒă‚·ăƒ„ăźæ©Ÿæą°ć­Šçż’PFI Seminar 2012/03/15 ă‚«ăƒŒăƒăƒ«ăšăƒăƒƒă‚·ăƒ„ăźæ©Ÿæą°ć­Šçż’
PFI Seminar 2012/03/15 ă‚«ăƒŒăƒăƒ«ăšăƒăƒƒă‚·ăƒ„ăźæ©Ÿæą°ć­Šçż’Preferred Networks
 
Naturaleza
NaturalezaNaturaleza
Naturalezadiosyely
 
Bumpless control for reduced thd in power factor correction circuits
Bumpless control for reduced thd in power factor correction circuitsBumpless control for reduced thd in power factor correction circuits
Bumpless control for reduced thd in power factor correction circuitsLeMeniz Infotech
 
April 13 2014 slideshow
April 13 2014 slideshowApril 13 2014 slideshow
April 13 2014 slideshowEarl Oswalt
 
Đ»ŃƒĐœĐœŃ‹Đ” Đ·Đ°Ń‚ĐŒĐ”ĐœĐžŃ
Đ»ŃƒĐœĐœŃ‹Đ” Đ·Đ°Ń‚ĐŒĐ”ĐœĐžŃĐ»ŃƒĐœĐœŃ‹Đ” Đ·Đ°Ń‚ĐŒĐ”ĐœĐžŃ
Đ»ŃƒĐœĐœŃ‹Đ” Đ·Đ°Ń‚ĐŒĐ”ĐœĐžŃEgor_2000
 
Google glass
Google glassGoogle glass
Google glassBK12741
 
Americans in Saint-Nazaire
Americans in Saint-NazaireAmericans in Saint-Nazaire
Americans in Saint-Nazairemdrouet44
 
НОВАЯ ВОЛНА КОМАНДНАЯ Đ­Đ€Đ€Đ•ĐšĐąĐ˜Đ’ĐĐžĐĄĐąĐŹ_Đ Đ°Đ·ĐČОтОД Đ›ĐžĐŽĐ”Ń€ĐŸĐČ ĐšĐŸŃƒŃ‡ĐžĐœĐł(1)
НОВАЯ ВОЛНА КОМАНДНАЯ Đ­Đ€Đ€Đ•ĐšĐąĐ˜Đ’ĐĐžĐĄĐąĐŹ_Đ Đ°Đ·ĐČОтОД Đ›ĐžĐŽĐ”Ń€ĐŸĐČ ĐšĐŸŃƒŃ‡ĐžĐœĐł(1)НОВАЯ ВОЛНА КОМАНДНАЯ Đ­Đ€Đ€Đ•ĐšĐąĐ˜Đ’ĐĐžĐĄĐąĐŹ_Đ Đ°Đ·ĐČОтОД Đ›ĐžĐŽĐ”Ń€ĐŸĐČ ĐšĐŸŃƒŃ‡ĐžĐœĐł(1)
НОВАЯ ВОЛНА КОМАНДНАЯ Đ­Đ€Đ€Đ•ĐšĐąĐ˜Đ’ĐĐžĐĄĐąĐŹ_Đ Đ°Đ·ĐČОтОД Đ›ĐžĐŽĐ”Ń€ĐŸĐČ ĐšĐŸŃƒŃ‡ĐžĐœĐł(1)Yelena Shaulova
 
Analysing Analytics: Evolution or Emperor's New Clothes? (Young OR Conference...
Analysing Analytics: Evolution or Emperor's New Clothes? (Young OR Conference...Analysing Analytics: Evolution or Emperor's New Clothes? (Young OR Conference...
Analysing Analytics: Evolution or Emperor's New Clothes? (Young OR Conference...Michael Mortenson
 
Camps gaëlle et lilou
Camps   gaëlle et lilouCamps   gaëlle et lilou
Camps gaëlle et liloumdrouet44
 

Andere mochten auch (12)

Search space reduction for holistic ligature recognition in Urdu Nastaliq scr...
Search space reduction for holistic ligature recognition in Urdu Nastaliq scr...Search space reduction for holistic ligature recognition in Urdu Nastaliq scr...
Search space reduction for holistic ligature recognition in Urdu Nastaliq scr...
 
Memory Networks, Neural Turing Machines, and Question Answering
Memory Networks, Neural Turing Machines, and Question AnsweringMemory Networks, Neural Turing Machines, and Question Answering
Memory Networks, Neural Turing Machines, and Question Answering
 
PFI Seminar 2012/03/15 ă‚«ăƒŒăƒăƒ«ăšăƒăƒƒă‚·ăƒ„ăźæ©Ÿæą°ć­Šçż’
PFI Seminar 2012/03/15 ă‚«ăƒŒăƒăƒ«ăšăƒăƒƒă‚·ăƒ„ăźæ©Ÿæą°ć­Šçż’PFI Seminar 2012/03/15 ă‚«ăƒŒăƒăƒ«ăšăƒăƒƒă‚·ăƒ„ăźæ©Ÿæą°ć­Šçż’
PFI Seminar 2012/03/15 ă‚«ăƒŒăƒăƒ«ăšăƒăƒƒă‚·ăƒ„ăźæ©Ÿæą°ć­Šçż’
 
Naturaleza
NaturalezaNaturaleza
Naturaleza
 
Bumpless control for reduced thd in power factor correction circuits
Bumpless control for reduced thd in power factor correction circuitsBumpless control for reduced thd in power factor correction circuits
Bumpless control for reduced thd in power factor correction circuits
 
April 13 2014 slideshow
April 13 2014 slideshowApril 13 2014 slideshow
April 13 2014 slideshow
 
Đ»ŃƒĐœĐœŃ‹Đ” Đ·Đ°Ń‚ĐŒĐ”ĐœĐžŃ
Đ»ŃƒĐœĐœŃ‹Đ” Đ·Đ°Ń‚ĐŒĐ”ĐœĐžŃĐ»ŃƒĐœĐœŃ‹Đ” Đ·Đ°Ń‚ĐŒĐ”ĐœĐžŃ
Đ»ŃƒĐœĐœŃ‹Đ” Đ·Đ°Ń‚ĐŒĐ”ĐœĐžŃ
 
Google glass
Google glassGoogle glass
Google glass
 
Americans in Saint-Nazaire
Americans in Saint-NazaireAmericans in Saint-Nazaire
Americans in Saint-Nazaire
 
НОВАЯ ВОЛНА КОМАНДНАЯ Đ­Đ€Đ€Đ•ĐšĐąĐ˜Đ’ĐĐžĐĄĐąĐŹ_Đ Đ°Đ·ĐČОтОД Đ›ĐžĐŽĐ”Ń€ĐŸĐČ ĐšĐŸŃƒŃ‡ĐžĐœĐł(1)
НОВАЯ ВОЛНА КОМАНДНАЯ Đ­Đ€Đ€Đ•ĐšĐąĐ˜Đ’ĐĐžĐĄĐąĐŹ_Đ Đ°Đ·ĐČОтОД Đ›ĐžĐŽĐ”Ń€ĐŸĐČ ĐšĐŸŃƒŃ‡ĐžĐœĐł(1)НОВАЯ ВОЛНА КОМАНДНАЯ Đ­Đ€Đ€Đ•ĐšĐąĐ˜Đ’ĐĐžĐĄĐąĐŹ_Đ Đ°Đ·ĐČОтОД Đ›ĐžĐŽĐ”Ń€ĐŸĐČ ĐšĐŸŃƒŃ‡ĐžĐœĐł(1)
НОВАЯ ВОЛНА КОМАНДНАЯ Đ­Đ€Đ€Đ•ĐšĐąĐ˜Đ’ĐĐžĐĄĐąĐŹ_Đ Đ°Đ·ĐČОтОД Đ›ĐžĐŽĐ”Ń€ĐŸĐČ ĐšĐŸŃƒŃ‡ĐžĐœĐł(1)
 
Analysing Analytics: Evolution or Emperor's New Clothes? (Young OR Conference...
Analysing Analytics: Evolution or Emperor's New Clothes? (Young OR Conference...Analysing Analytics: Evolution or Emperor's New Clothes? (Young OR Conference...
Analysing Analytics: Evolution or Emperor's New Clothes? (Young OR Conference...
 
Camps gaëlle et lilou
Camps   gaëlle et lilouCamps   gaëlle et lilou
Camps gaëlle et lilou
 

Ähnlich wie 20140327 - Hashing Object Embedding

Duplicate Detection on Hoaxy Dataset
Duplicate Detection on Hoaxy DatasetDuplicate Detection on Hoaxy Dataset
Duplicate Detection on Hoaxy DatasetSameera Horawalavithana
 
Xin Yao: "What can evolutionary computation do for you?"
Xin Yao: "What can evolutionary computation do for you?"Xin Yao: "What can evolutionary computation do for you?"
Xin Yao: "What can evolutionary computation do for you?"ieee_cis_cyprus
 
HyperQA: A Framework for Complex Question-Answering
HyperQA: A Framework for Complex Question-AnsweringHyperQA: A Framework for Complex Question-Answering
HyperQA: A Framework for Complex Question-AnsweringJinho Choi
 
Deep learning 1.0 and Beyond, Part 1
Deep learning 1.0 and Beyond, Part 1Deep learning 1.0 and Beyond, Part 1
Deep learning 1.0 and Beyond, Part 1Deakin University
 
Hala skafkeynote@conferencedata2021
Hala skafkeynote@conferencedata2021Hala skafkeynote@conferencedata2021
Hala skafkeynote@conferencedata2021hala Skaf
 
Deep Learning Explained
Deep Learning ExplainedDeep Learning Explained
Deep Learning ExplainedMelanie Swan
 
Similarity-preserving hash for content-based audio retrieval using unsupervis...
Similarity-preserving hash for content-based audio retrieval using unsupervis...Similarity-preserving hash for content-based audio retrieval using unsupervis...
Similarity-preserving hash for content-based audio retrieval using unsupervis...IJECEIAES
 
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!Sri Ambati
 
Deep learning and reasoning: Recent advances
Deep learning and reasoning: Recent advancesDeep learning and reasoning: Recent advances
Deep learning and reasoning: Recent advancesDeakin University
 
Interlinking educational data to Web of Data (Thesis presentation)
Interlinking educational data to Web of Data (Thesis presentation)Interlinking educational data to Web of Data (Thesis presentation)
Interlinking educational data to Web of Data (Thesis presentation)Enayat Rajabi
 
Effective data mining for proper
Effective data mining for properEffective data mining for proper
Effective data mining for properIJDKP
 
User Interfaces that Design Themselves: Talk given at Data-Driven Design Day ...
User Interfaces that Design Themselves: Talk given at Data-Driven Design Day ...User Interfaces that Design Themselves: Talk given at Data-Driven Design Day ...
User Interfaces that Design Themselves: Talk given at Data-Driven Design Day ...Aalto University
 
EDBT 2015: Summer School Overview
EDBT 2015: Summer School OverviewEDBT 2015: Summer School Overview
EDBT 2015: Summer School Overviewdgarijo
 
Hybrid Meta-Heuristic Algorithms For Solving Network Design Problem
Hybrid Meta-Heuristic Algorithms For Solving Network Design ProblemHybrid Meta-Heuristic Algorithms For Solving Network Design Problem
Hybrid Meta-Heuristic Algorithms For Solving Network Design ProblemAlana Cartwright
 
Linked Open Data (LOD) part 1
Linked Open Data (LOD) part 1Linked Open Data (LOD) part 1
Linked Open Data (LOD) part 1IPLODProject
 
In tech application-of_data_mining_technology_on_e_learning_material_recommen...
In tech application-of_data_mining_technology_on_e_learning_material_recommen...In tech application-of_data_mining_technology_on_e_learning_material_recommen...
In tech application-of_data_mining_technology_on_e_learning_material_recommen...Enhmandah Hemeelee
 

Ähnlich wie 20140327 - Hashing Object Embedding (20)

Duplicate Detection on Hoaxy Dataset
Duplicate Detection on Hoaxy DatasetDuplicate Detection on Hoaxy Dataset
Duplicate Detection on Hoaxy Dataset
 
Xin Yao: "What can evolutionary computation do for you?"
Xin Yao: "What can evolutionary computation do for you?"Xin Yao: "What can evolutionary computation do for you?"
Xin Yao: "What can evolutionary computation do for you?"
 
HyperQA: A Framework for Complex Question-Answering
HyperQA: A Framework for Complex Question-AnsweringHyperQA: A Framework for Complex Question-Answering
HyperQA: A Framework for Complex Question-Answering
 
Deep learning 1.0 and Beyond, Part 1
Deep learning 1.0 and Beyond, Part 1Deep learning 1.0 and Beyond, Part 1
Deep learning 1.0 and Beyond, Part 1
 
Hala skafkeynote@conferencedata2021
Hala skafkeynote@conferencedata2021Hala skafkeynote@conferencedata2021
Hala skafkeynote@conferencedata2021
 
Deep Learning Explained
Deep Learning ExplainedDeep Learning Explained
Deep Learning Explained
 
Similarity-preserving hash for content-based audio retrieval using unsupervis...
Similarity-preserving hash for content-based audio retrieval using unsupervis...Similarity-preserving hash for content-based audio retrieval using unsupervis...
Similarity-preserving hash for content-based audio retrieval using unsupervis...
 
JOSA TechTalks - Machine Learning in Practice
JOSA TechTalks - Machine Learning in PracticeJOSA TechTalks - Machine Learning in Practice
JOSA TechTalks - Machine Learning in Practice
 
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
 
Deep learning and reasoning: Recent advances
Deep learning and reasoning: Recent advancesDeep learning and reasoning: Recent advances
Deep learning and reasoning: Recent advances
 
Interlinking educational data to Web of Data (Thesis presentation)
Interlinking educational data to Web of Data (Thesis presentation)Interlinking educational data to Web of Data (Thesis presentation)
Interlinking educational data to Web of Data (Thesis presentation)
 
Effective data mining for proper
Effective data mining for properEffective data mining for proper
Effective data mining for proper
 
AI Science
AI Science AI Science
AI Science
 
ME Synopsis
ME SynopsisME Synopsis
ME Synopsis
 
User Interfaces that Design Themselves: Talk given at Data-Driven Design Day ...
User Interfaces that Design Themselves: Talk given at Data-Driven Design Day ...User Interfaces that Design Themselves: Talk given at Data-Driven Design Day ...
User Interfaces that Design Themselves: Talk given at Data-Driven Design Day ...
 
EDBT 2015: Summer School Overview
EDBT 2015: Summer School OverviewEDBT 2015: Summer School Overview
EDBT 2015: Summer School Overview
 
Deep Learning 2.0
Deep Learning 2.0Deep Learning 2.0
Deep Learning 2.0
 
Hybrid Meta-Heuristic Algorithms For Solving Network Design Problem
Hybrid Meta-Heuristic Algorithms For Solving Network Design ProblemHybrid Meta-Heuristic Algorithms For Solving Network Design Problem
Hybrid Meta-Heuristic Algorithms For Solving Network Design Problem
 
Linked Open Data (LOD) part 1
Linked Open Data (LOD) part 1Linked Open Data (LOD) part 1
Linked Open Data (LOD) part 1
 
In tech application-of_data_mining_technology_on_e_learning_material_recommen...
In tech application-of_data_mining_technology_on_e_learning_material_recommen...In tech application-of_data_mining_technology_on_e_learning_material_recommen...
In tech application-of_data_mining_technology_on_e_learning_material_recommen...
 

KĂŒrzlich hochgeladen

Just Call Vip call girls Bellary Escorts ☎9352988975 Two shot with one girl ...
Just Call Vip call girls Bellary Escorts ☎9352988975 Two shot with one girl ...Just Call Vip call girls Bellary Escorts ☎9352988975 Two shot with one girl ...
Just Call Vip call girls Bellary Escorts ☎9352988975 Two shot with one girl ...gajnagarg
 
âž„đŸ” 7737669865 đŸ”â–» Dindigul Call-girls in Women Seeking Men 🔝Dindigul🔝 Escor...
âž„đŸ” 7737669865 đŸ”â–» Dindigul Call-girls in Women Seeking Men  🔝Dindigul🔝   Escor...âž„đŸ” 7737669865 đŸ”â–» Dindigul Call-girls in Women Seeking Men  🔝Dindigul🔝   Escor...
âž„đŸ” 7737669865 đŸ”â–» Dindigul Call-girls in Women Seeking Men 🔝Dindigul🔝 Escor...amitlee9823
 
Discover Why Less is More in B2B Research
Discover Why Less is More in B2B ResearchDiscover Why Less is More in B2B Research
Discover Why Less is More in B2B Researchmichael115558
 
Call Girls Jalahalli Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
Call Girls Jalahalli Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...Call Girls Jalahalli Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
Call Girls Jalahalli Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...amitlee9823
 
Call Girls Indiranagar Just Call 👗 9155563397 👗 Top Class Call Girl Service B...
Call Girls Indiranagar Just Call 👗 9155563397 👗 Top Class Call Girl Service B...Call Girls Indiranagar Just Call 👗 9155563397 👗 Top Class Call Girl Service B...
Call Girls Indiranagar Just Call 👗 9155563397 👗 Top Class Call Girl Service B...only4webmaster01
 
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24  Building Real-Time Pipelines With FLaNKDATA SUMMIT 24  Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNKTimothy Spann
 
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...ZurliaSoop
 
Call Girls In Nandini Layout ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Nandini Layout ☎ 7737669865 đŸ„” Book Your One night StandCall Girls In Nandini Layout ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Nandini Layout ☎ 7737669865 đŸ„” Book Your One night Standamitlee9823
 
âž„đŸ” 7737669865 đŸ”â–» mahisagar Call-girls in Women Seeking Men 🔝mahisagar🔝 Esc...
âž„đŸ” 7737669865 đŸ”â–» mahisagar Call-girls in Women Seeking Men  🔝mahisagar🔝   Esc...âž„đŸ” 7737669865 đŸ”â–» mahisagar Call-girls in Women Seeking Men  🔝mahisagar🔝   Esc...
âž„đŸ” 7737669865 đŸ”â–» mahisagar Call-girls in Women Seeking Men 🔝mahisagar🔝 Esc...amitlee9823
 
CHEAP Call Girls in Rabindra Nagar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Rabindra Nagar  (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICECHEAP Call Girls in Rabindra Nagar  (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Rabindra Nagar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE9953056974 Low Rate Call Girls In Saket, Delhi NCR
 
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...amitlee9823
 
VIP Model Call Girls Hinjewadi ( Pune ) Call ON 8005736733 Starting From 5K t...
VIP Model Call Girls Hinjewadi ( Pune ) Call ON 8005736733 Starting From 5K t...VIP Model Call Girls Hinjewadi ( Pune ) Call ON 8005736733 Starting From 5K t...
VIP Model Call Girls Hinjewadi ( Pune ) Call ON 8005736733 Starting From 5K t...SUHANI PANDEY
 
Call Girls In Attibele ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Attibele ☎ 7737669865 đŸ„” Book Your One night StandCall Girls In Attibele ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Attibele ☎ 7737669865 đŸ„” Book Your One night Standamitlee9823
 
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...amitlee9823
 
Call Girls In Bellandur ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Bellandur ☎ 7737669865 đŸ„” Book Your One night StandCall Girls In Bellandur ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Bellandur ☎ 7737669865 đŸ„” Book Your One night Standamitlee9823
 
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...amitlee9823
 
Detecting Credit Card Fraud: A Machine Learning Approach
Detecting Credit Card Fraud: A Machine Learning ApproachDetecting Credit Card Fraud: A Machine Learning Approach
Detecting Credit Card Fraud: A Machine Learning ApproachBoston Institute of Analytics
 

KĂŒrzlich hochgeladen (20)

Just Call Vip call girls Bellary Escorts ☎9352988975 Two shot with one girl ...
Just Call Vip call girls Bellary Escorts ☎9352988975 Two shot with one girl ...Just Call Vip call girls Bellary Escorts ☎9352988975 Two shot with one girl ...
Just Call Vip call girls Bellary Escorts ☎9352988975 Two shot with one girl ...
 
âž„đŸ” 7737669865 đŸ”â–» Dindigul Call-girls in Women Seeking Men 🔝Dindigul🔝 Escor...
âž„đŸ” 7737669865 đŸ”â–» Dindigul Call-girls in Women Seeking Men  🔝Dindigul🔝   Escor...âž„đŸ” 7737669865 đŸ”â–» Dindigul Call-girls in Women Seeking Men  🔝Dindigul🔝   Escor...
âž„đŸ” 7737669865 đŸ”â–» Dindigul Call-girls in Women Seeking Men 🔝Dindigul🔝 Escor...
 
Discover Why Less is More in B2B Research
Discover Why Less is More in B2B ResearchDiscover Why Less is More in B2B Research
Discover Why Less is More in B2B Research
 
Call Girls Jalahalli Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
Call Girls Jalahalli Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...Call Girls Jalahalli Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
Call Girls Jalahalli Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
 
Call Girls Indiranagar Just Call 👗 9155563397 👗 Top Class Call Girl Service B...
Call Girls Indiranagar Just Call 👗 9155563397 👗 Top Class Call Girl Service B...Call Girls Indiranagar Just Call 👗 9155563397 👗 Top Class Call Girl Service B...
Call Girls Indiranagar Just Call 👗 9155563397 👗 Top Class Call Girl Service B...
 
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24  Building Real-Time Pipelines With FLaNKDATA SUMMIT 24  Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
 
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
 
Call Girls In Nandini Layout ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Nandini Layout ☎ 7737669865 đŸ„” Book Your One night StandCall Girls In Nandini Layout ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Nandini Layout ☎ 7737669865 đŸ„” Book Your One night Stand
 
âž„đŸ” 7737669865 đŸ”â–» mahisagar Call-girls in Women Seeking Men 🔝mahisagar🔝 Esc...
âž„đŸ” 7737669865 đŸ”â–» mahisagar Call-girls in Women Seeking Men  🔝mahisagar🔝   Esc...âž„đŸ” 7737669865 đŸ”â–» mahisagar Call-girls in Women Seeking Men  🔝mahisagar🔝   Esc...
âž„đŸ” 7737669865 đŸ”â–» mahisagar Call-girls in Women Seeking Men 🔝mahisagar🔝 Esc...
 
CHEAP Call Girls in Rabindra Nagar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Rabindra Nagar  (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICECHEAP Call Girls in Rabindra Nagar  (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Rabindra Nagar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
 
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
 
VIP Model Call Girls Hinjewadi ( Pune ) Call ON 8005736733 Starting From 5K t...
VIP Model Call Girls Hinjewadi ( Pune ) Call ON 8005736733 Starting From 5K t...VIP Model Call Girls Hinjewadi ( Pune ) Call ON 8005736733 Starting From 5K t...
VIP Model Call Girls Hinjewadi ( Pune ) Call ON 8005736733 Starting From 5K t...
 
Call Girls In Attibele ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Attibele ☎ 7737669865 đŸ„” Book Your One night StandCall Girls In Attibele ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Attibele ☎ 7737669865 đŸ„” Book Your One night Stand
 
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
 
Call Girls In Bellandur ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Bellandur ☎ 7737669865 đŸ„” Book Your One night StandCall Girls In Bellandur ☎ 7737669865 đŸ„” Book Your One night Stand
Call Girls In Bellandur ☎ 7737669865 đŸ„” Book Your One night Stand
 
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
 
Abortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Jeddah | +966572737505 | Get CytotecAbortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Jeddah | +966572737505 | Get Cytotec
 
Abortion pills in Doha Qatar (+966572737505 ! Get Cytotec
Abortion pills in Doha Qatar (+966572737505 ! Get CytotecAbortion pills in Doha Qatar (+966572737505 ! Get Cytotec
Abortion pills in Doha Qatar (+966572737505 ! Get Cytotec
 
Detecting Credit Card Fraud: A Machine Learning Approach
Detecting Credit Card Fraud: A Machine Learning ApproachDetecting Credit Card Fraud: A Machine Learning Approach
Detecting Credit Card Fraud: A Machine Learning Approach
 
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts ServiceCall Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
 

20140327 - Hashing Object Embedding

  • 1. 1 Hashing: Object Embedding Reporter: Xu Jiaming (PH.D Student) Date: 2014.03.27 Computational-Brain Research Center Institute of Automation, Chinese Academy of Sciences Report
  • 2. 2 First, What is Embedding? [ ć‡șè‡Ș ]: https://en.wikipedia.org/wiki/Embedding When some object X is said to be embedded in another object Y, the embedding is given by some injective and structure- preserving map f : X → Y. The precise meaning of "structure- preserving" depends on the kind of mathematical structure of which X and Y are instances. Structure-Preserving in IR: 1 2 1 2 : (X ,X ) (Y ,Y ) f Sim Sim → ≈ X Y
  • 3. 3 Then, What is Hash? [ ć‡șè‡Ș ]: https://en.wikipedia.org/wiki/Hash_table The hash function will assign each key to a unique bucket, but this situation is rarely achievable in practice (usually some keys will hash to the same bucket). Instead, most hash table designs assume that hash collisions—different keys that are assigned by the hash function to the same bucket—will occur and must be accommodated in some way.
  • 4. 4 Combine the Two Properties [1998, Piotr Indyk, cited: 1847] Locality Sensitive Hashing . ( , ) , . Pr[ ( ) ( )] 1 . ( , ) (1 ), . Pr[ ( ) ( )] 1 if D p q r then h p h q p if D p q r then h p h q pΔ ≀ = ≄ > + = ≄
  • 5. 5 Overview of Hashing Real World Binary Space 2000 values 32 bits Binary Reduction
  • 7. 7 Learning to Hash Data-Oblivious Data-Aware Description Methods LSI, RBM, SpH, STH, 
 LSH, Kernel-LSH, SimHash, 

  • 8. 8 Data-Oblivious: SimHash [WWW.2007] Text 
 
 Observed Features W1 W2 Wn 100110 W1 110000 W2 001001 Wn 
 
 W1 –W1 -W1 W1 W1 -W1 W2 W2 -W2 -W2 -W2 -W2 -Wn –Wn Wn –Wn –Wn Wn 
 
13, 108, -22, -5, -32, 551, 1, 0, 0, 0, 1 Step1: Compute TF- IDF Step2: Hash Function Step3: Signature Step4: Sum Step5: Generate Fingerprint
  • 9. 9 Data-Aware: Spectral Hashing [NIPS.2008] 2 min : . .: { 1,1} 0 1 ij i j ij k i i i T i i i S y y s t y y y y n − ∈ − = = ∑ ∑ ∑ I min : ( ( ) ) . .: ( , ) { 1,1} 0 T k T T trace Y D W Y s t Y i j − ∈ − = = Y 1 Y Y I Laplacian Eigenmap XW Y=
  • 10. 10 Some Questions? 1. Can we obtain hashing codes by binarizing the real-valued low- dimensional vectors such as LSI? 2. Can we get hashing codes by Deep Learning approaches such as RBM, or AutoEncoder?
  • 11. 11 Some Questions? 1. Can we obtain hashing codes by binarizing the real-valued low- dimensional vectors such as LSI? Of Course ! [R. Salakhutdinov, G. Hinton. Semantic Hashing, SIGIR2007] 2. Can we get hashing codes by Deep Learning approaches such as RBM, or AutoEncoder? No Problem ! [R. Salakhutdinov, G. Hinton. Semantic Hashing, SIGIR2007]
  • 12. 12 In 2013, What Did They Think About? Total: 30
  • 13. 13 1/9 - ICML2013: Title: Learning Hash Functions Using Column Generation Authors: Xi Li, Guosheng Lin, Chunhua Shen, Anton van den Hengel, Anthony Dick Organization: The University of Adelaide (Australia) Based On: NIPS2005: Distance Metric Learning for Large Margin Nearest Neighbor ClassiïŹcation Motivation: In content based image retrieval, to collect feedback, users may be required to report whether image x looks more similar to x+ than it is to a third image x−. This task is typically much easier than to label each individual image. 11 min . . 0, 0; ( , ) ( , ) 1 , J i i H i i H i i i C s t d d i Ο Ο Ο = − + + ≄ ≄ − ≄ − ∀ ∑w,Ο w w x x x x
  • 14. 14 2/9 - ICML2013: Title: Predictable Dual-View Hashing Authors: Mohammad Rastegari, Jonghyun Choi, Shobeir Fakhraei, Hal Daume III, Larry S. Davis Organization: The University of Maryland (USA) Motivation: It is often the case that information about data are available from two or more views, e.g., images and their textual descriptions. It is highly desirable to embed information from both domains in the binary codes, to increase search and retrieval capabilities. 2 2 2 2 2 2 2 2 min . . sgn(W X ) sgn(W X ) T T T T V V T T T T T V V V T T T T T V V V W X Y Y Y I W X Y Y Y I s t Y Y − + − + − + − = = Y,U
  • 15. 15 3/9 - SIGIR2013: Title: Semantic Hashing Using Tags and Topic Modeling. Authors: Qifan Wang, Dan Zhang, Luo Si Organization: Purdue University (USA) Motivation: Two major issues are not addressed in the existing hashing methods: (1) Tag information is not fully utilized in previous methods. Most existing methods only deal with the contents of documents without utilizing the information contained in tags; (2) Document similarity in the original keyword feature space is used as guidance for generating hashing codes in previous methods, which may not fully reflect the semantic relationship. 1 2 2 2 2 min ( ) . . { 1,1} , 0 T F k n C s t Îł × − + + − ∈ − = Y,U T U Y U YΞ Y Y1 g
  • 16. 16 3/9 - SIGIR2013: Title: Semantic Hashing Using Tags and Topic Modeling. Authors: Qifan Wang, Dan Zhang, Luo Si Organization: Purdue University (USA) Motivation: Two major issues are not addressed in the existing hashing methods: (1) Tag information is not fully utilized in previous methods. Most existing methods only deal with the contents of documents without utilizing the information contained in tags; (2) Document similarity in the original keyword feature space is used as guidance for generating hashing codes in previous methods, which may not fully reflect the semantic relationship. 1 2 2 2 2 min ( ) . . { 1,1} , 0 T F k n C s t Îł × − + + − ∈ − = Y,U T U Y U YΞ Y Y1 g Our experiments on 20Newsgroups
  • 17. 17 4/9 - IJCAI2013: Title: A Unified Approximate Nearest Neighbor Search Scheme by Combining Data Structure and Hashing. Authors: Debing Zhang, Genmao Yang, Yao Hu, Zhongming Jin, Deng Cai, Xiaofei He Organization: Zhejiang University (China) Motivation: Traditionally, to solve problem of nearest neighbor search, researchers mainly focus on building effective data structures such as hierarchical k-means tree or using hashing methods to accelerate the query process. In this paper, we propose a novel unified approximate nearest neighbor search scheme to combine the advantages of both the effective data structure and the fast Hamming distance computation in hashing methods.
  • 18. 18 5/9 - CVPR2013: Title: K-means Hashing: an Affinity-Preserving Quantization Method for Learning Binary Compact Codes. Authors: Kaiming He, Fang Wen, Jian Sun Organization: Microsoft Research Asia (China) Motivation: Both Hamming-based methods and lookup-based methods are of growing interest recently, and each category has its benefits depending on the scenarios. The lookup-based methods have been shown more accurate than some Hamming methods with the same code-length. However, the lookup-based distance computation is slower than the Hamming distance computation. Hamming methods also have the advantage that the distance computation is problem-independent 1 1 2 0 0 ( ( , ) ( , )) k k aff ij i j h i j E w d c c d i j − − = = = −∑∑
  • 19. 19 6/9 - ICCV2013: Title: Complementary Projection Hashing. Authors: Zhongming Jin1 , Yao Hu1 , Yue Lin1 , Debing Zhang1 , Shiding Lin2 , Deng Cai1 , Xuelong Li3 Organization: 1. Zhejiang University, 2. Baidu Inc., 3. Chinese Academy of Sciences, Xi’an (China) Motivation: 1. (a) The hyperplane a crosses the sparse region and the neighbors are quantized into the same bucket; (b) The hyperplane b crosses the dense region and the neighbors are quantized into the different buckets. Apparently, the hyperplane a is more suitable as a hashing function. 2. (a) (b) Both the hyperplane a and the hyperplane b can evenly separated the data. (c) However, putting them together does not generate a good two bits hash function. (d) A better example for two bits hash function
  • 20. 20 7/9 - CVPR2013: Title: Hash Bit Selection: a Unified Solution for Selection Problems in Hashing. Authors: Xianglong Liu1 , Junfeng He2,3 , Bo Lang1 , Shih-Fu Chang2 . Organization: 1. Beihang University(China), 2. Columbia University(US), 3. Facebook(US) Motivation: Recent years have witnessed the active development of hashing techniques for nearest neighbor search over big datasets. However, to apply hashing techniques successfully, there are several important issues remaining open in selection features, hashing algorithms, parameter settings  kernels, etc.
  • 21. 21 8/9 - ICCV2013: Title: A General Two-Step Approach to Learning-Based Hashing. Authors: Guosheng Lin, Chunhua Shen, David Suter, Anton van den Hengel. Organization: University of Adelaide (Australia) Based On: SIGIR2010: Self-Taught Hashing for Fast Similarity Search Motivation: Most existing approaches to hashing apply a single form of hash function, and an optimization process which is typically deeply coupled to this specific form. This tight coupling restricts the flexibility of the method to respond to the data, and can result in complex optimization problems that are difficult to solve. Their framework decomposes hashing learning problem into two steps: hash bit learning and hash function learning based on the learned bits.
  • 22. 22 9/9 - IJCAI2013: Title: Smart Hashing Update for Fast Response. Authors: Qiang Yang, Long-Kai Huang, Wei-Shi Zheng, Yingbiao Ling. Organization: Sun Yat-sen University (China) Based On: DMKD2012: Active Hashing and Its Application to Image and Text Retrieval Motivation: Although most existing hashing-based methods have been proven to obtain high accuracy, they are regarded as passive hashing and assume that the labeled points are provided in advance. In this paper, they consider updating a hashing model upon gradually increased labeled data in a fast response to users, called smart hashing update (SHU). 1. Consistency-based Selection; 2. Similarity-based Selection. [CVPR.2012] ( , ) min{ ( , , 1), ( , ,1)}Diff k j num k j num k j= − 2 { 1,1} 1 min l r l T l l H F Q H H S r× ∈ − = − 2 1 1 {1,2,...,r} min k k T k r r Fk R rS H H− − ∈ = −
  • 23. 23 Reporter: Xu Jiaming (Ph.D Student) Date: 2014.03.27 Computational-Brain Research Center Institute of Automation, Chinese Academy of Sciences

Hinweis der Redaktion

  1. ć€§ćź¶ć„œïŒŒæˆ‘ä»Šć€©ç»™ć€§ćź¶ćšçš„æŠ„ć‘Šæ˜ŻHashingïŒŒæˆ‘èż™é‡Œç»™ć‡ș侀äžȘć‰Żæ ‡éą˜ïŒŒObject EmbeddingïŒŒć…¶ćźžäčŸæ˜Żć“—äŒ—ć–ćź ïŒŒć› äžșWord Embeddingèż™äžȘèŻæŻ”èŸƒæ—¶é«Šć˜›ă€‚
  2. æˆ‘ä»Źć…ˆæ„ć…œäžȘćœˆïŒŒä»€äčˆć«EmbeddingWikiPediaäžŠèż™äčˆè§Łé‡Šçš„ïŒŒćŠ‚æžœèŻŽäž€äžȘ物䜓Xèą«Embbedingćˆ°ç‰©äœ“YäžŠïŒŒé‚Łäčˆèż™äžȘEmbeddingèż‡çš‹é€šćžžæ˜Żç”±äž€äș›ć•ć°„äž”ç»“æž„äżć­˜çš„æ˜ ć°„ć‡œæ•°ćźŒæˆă€‚ć…łäșŽç»“æž„äżć­˜çš„çČŸçĄźćźšäč‰ć–憳äșŽX,Yèż™äž€äžȘćźžäœ“ć…·æœ‰ä»€äčˆæ ·çš„æ•°ć­Šç»“æž„ă€‚çź€ć•ç†è§ŁïŒŒć…¶ćźžEmbeddingć°±æ˜Żäž€ç§ç»“æž„æ€§äżć­˜çš„æ˜ ć°„ć…łçł»ïŒŒć€§ć€šæ•°æ„èŻŽéƒœæ˜Żé™ç»ŽïŒŒæˆ‘ä»Źæ‰€ç†ŸçŸ„çš„ć€§ć€šæ•°çš„é™ç»Žćˆ†è§Łæ–čæł•ïŒŒćŠ‚SVDPCALDAéƒœæ˜Żäž€ç§Embeding。朹IRäž­ïŒŒç»“æž„äżć­˜æ€§ć°±æ˜ŻćžŒæœ›é€šèż‡äž€ç§æ˜ ć°„ć…łçł»ćŽïŒŒćŽŸç©ș闎䞭物䜓äč‹é—Žçš„ç›žäŒŒæ€§èƒœćœšç›źæ ‡ç©ș问侭ćŸșæœŹäżæŒäžć˜ă€‚
  3. ç„¶ćŽæˆ‘ä»Źæ„çœ‹ä»€äčˆć«HashćŻčäșŽć†™ä»Łç çš„äșșæ„èŻŽïŒŒHashæœ€ç†Ÿæ‚‰äžèż‡ïŒŒæ˜Żäž€ç§ć…žćž‹çš„Key-Valueç»“æž„ïŒŒæœ€ćžžè§çš„çź—æł•èŽ«èż‡äșŽMD5ă€‚ć…¶èźŸèźĄæ€æƒłæ˜ŻäœżKeyé›†ćˆäž­çš„ä»»æ„ć…łé”źć­—èƒœć€Ÿć°œćŻèƒœć‡ćŒ€çš„ć˜æąćˆ°Valueç©șé—Žäž­ïŒŒäžćŒçš„KeyćŻčćș”äžćŒçš„ValueïŒŒćłäœżKeyć€ŒćȘæœ‰èœ»ćŸźć˜ćŒ–ïŒŒValueć€ŒäčŸäŒšć‘ç”ŸćŸˆć€§ćœ°ć˜ćŒ–ă€‚WipikPediaäžŠèż™äčˆè§Łé‡ŠèŻŽïŒŒHashć‡œæ•°é€šćžžæ˜Żç»™æŻKeyć€Œćˆ†é…äž€äžȘç‹Źäž€æ— äșŒçš„æĄ¶ă€‚ćŒæ—¶äčŸæŒ‡æ˜Žäș†ćœšćźžé™…ćș”ç”šäž­ïŒŒèż™ç§æƒ…ć†”æ˜ŻéšŸä»„ćźžçŽ°çš„ïŒŒæ€»æœ‰äž€äș›Keyć€ŒäŒšćˆ†ćˆ°ćŒäž€äžȘæĄ¶ć†…ă€‚èż™ç§æƒ…ć†”èą«ćźšäč‰äžșHash CollisionsHashçą°æ’žïŒŒèż™ćœšHashć‡œæ•°èźŸèźĄäž­èŠć°œćŻèƒœéżć…ă€‚Hashèą«ćčżæł›ćș”甚äčŸæ˜Żć› äžșćźƒæ€„ćż«çš„Keyć€ŒæŁ€çŽąé€ŸćșŠïŒŒç†æƒłæƒ…ć†”äž‹ïŒŒæ—¶è€—æ˜ŻO(1)ă€‚äœ†æ˜Żèż™ç§æ•Łćˆ—çš„Hashć‡œæ•°æ˜Ÿç„¶ćč¶äžæ»Ąè¶łEmbeddingçš„æ€§èŽšïŒŒć› äžșæČĄæœ‰äżć­˜ç‰©äœ“æ‰€ć›șæœ‰çš„ç»“æž„æ€§èŽšă€‚
  4. 那äčˆæˆ‘ä»ŹäŒšæƒłïŒŒèƒœćŠèźŸèźĄèż™æ ·äž€ç§HashïŒšćźƒäœżç›žäŒŒKeyć€ŒèźĄçź—ć‡ș的Valueć€Œç›žćŒæˆ–ćœšæŸç§ćșŠé‡äž‹ç›žèż‘ïŒŒç”šè‡łćŸ—ćˆ°çš„Valueć€Œèƒœć€Ÿäżç•™ç‰©äœ“ćŽŸć§‹ç©șé—Žçš„äżĄæŻïŒŒèż™æ ·ç›žćŒæˆ–ç›žèż‘çš„æ–‡ä»¶èƒœć€Ÿä»„Hash的æ–čćŒèą«ćż«é€ŸæŁ€çŽąć‡șæ„ïŒŒæˆ–ç”šäœœćż«é€Ÿçš„ç›žäŒŒæ€§æŻ”ćŻč。98ćčŽIndyk朹StanfordèŻ»PHDæ—¶äžŽä»–çš„ćŻŒćžˆäž€è”·æć‡ș的侀种hashæ–čæł•:äœçœźæ•æ„Ÿć“ˆćžŒïŒˆLocal Sensitive Hashing, LSHïŒ‰æ­Łć„œæ»Ąè¶łäș†èż™ç§éœ€æ±‚。ćŻčäșŽä»»æ„çš„q,p汞äșŽSç©șé—ŽïŒŒè‹„ä»Žé›†ćˆS戰U的Hashć‡œæ•°h,ćŻčè·çŠ»ć‡œæ•°D(p,q)ïŒŒæ»Ąè¶łć…ŹćŒäž­çš„æĄä»¶ćˆ™ç§°D(p,q)æ˜Żäœçœźæ•æ„Ÿçš„ă€‚ć…¶ćźžć°±æ˜ŻèŻŽïŒŒćŽŸç©șé—Žè·çŠ»èż‘çš„æ˜ ć°„ćˆ°ćŒäž€äžȘæĄ¶äž­çš„æŠ‚çŽ‡ć°±æŻ”èŸƒć€§ă€‚
  5. çŽ°ćœšçź€ć•æ„çœ‹ïŒŒHashingć…¶ćźžć°±æ˜Żçź€ć•ćšäș†äž€ä»¶äș‹ïŒŒćŻč高绎ç©șé—Žçš„ç‰©äœ“èż›èĄŒé™ç»ŽäșŒć€ŒćŒ–ïŒŒćŒæ—¶äżć­˜ćŽŸç©șé—Žäž­çš„ç›žäŒŒæ€§äżĄæŻă€‚èż™ç§äșŒèż›ćˆ¶äœŽç»Žç‰čćŸé‡èĄšç€șæ–čæł•ćœšäżĄæŻæŁ€çŽąéą†ćŸŸæ˜ŻćŸˆæœ‰ç”šçš„ïŒŒçŽ°ćœšç§»ćŠšç»ˆç«Żçš„ćș”ç”šćŸˆç«ïŒŒćŠ‚æžœæˆ‘ä»Źç”šæ‰‹æœșæ‹äž€ćŒ ç…§ç‰‡èż›èĄŒäżĄæŻæŁ€çŽąçš„èŻïŒŒç»™æœćŠĄć™šäžŠäŒ äž€äžȘ32bitsçš„äżĄæŻèŠæŻ”äžŠäŒ äž€ćŒ ć›Ÿç‰‡ćż«æ·çš„ć€šïŒŒè€Œäž”ćœšæœćŠĄć™šç«Żèż›èĄŒäœèżçź—èŠæŻ”ćźžć€Œèżçź—ćż«æ·çš„ć€šă€‚çŽ°ćœšć€§æ•°æźæ—¶ä»ŁïŒŒHashingæ–čæł•ć°±ć˜ćŸ—æ›ŽćŠ é‡èŠïŒŒçŽ°ćœšćŸźèœŻć’Œç™ŸćșŠçš„æ”·é‡äżĄæŻç»„ćœšćș”ćŻčć€§æ•°æźæ–čéąéƒœćœšćšHashingæ–čéąçš„ç ”ç©¶ă€‚
  6. Hashingæ–čæł•èˆçČŸæ±‚ćż«çš„æ–čæł•äčŸç»™æˆ‘ä»Źäž€ç§éąćŻčć€§æ•°æźçš„æ€ćșŠć°±æ˜ŻApproximation-é€Œèż‘ă€‚
  7. Hashingæ–čæł•ćŻćˆ†äžșäž€ç§ïŒŒäž€ç§æ˜ŻćŸșäșŽæ•°æźé›†çš„ïŒŒäž€ç§æ˜ŻäžŽæ•°æźæ— ć…łçš„ă€‚æˆ‘ä»Źćˆ†ćˆ«ä»‹ç»äž€ç§æ–čæł•ă€‚
  8. äžŽæ•°æźæ— ć…łçš„çź—æł•ć€§ć€šæ˜Żéšćłæ˜ ć°„äș§ç”Ÿçš„ïŒŒäŸ‹ćŠ‚WWW2007侊Googleć‘èĄšçš„äž€çŻ‡ć…łäșŽSimHashçš„æ–‡ç« ïŒŒćźƒæŠŠäž€äžȘæ–‡æœŹæ‹†æˆnäžȘToken,ç„¶ćŽé€šèż‡Hashć‡œæ•°æ˜ ć°„ćˆ°ć€šäžȘćčłéąäžŠïŒŒèż™äžȘHashFunctionćč¶äžć…·æœ‰ç›žäŒŒäżć­˜æ€§ïŒŒæŒ‰äœæ±‚ć’Œç„¶ćŽäșŒć€ŒćŒ–ć°±ćŸ—ćˆ°äș†äž€äžȘHash指çșč敎äžȘçź—æł•éžćžžçź€ć•ïŒŒéšćłæ˜ ć°„ćŻä»„ç†è§Łäžșćœšé«˜ç»Žç©șé—Žäž­èż›èĄŒćˆ‡ć‰ČïŒŒćœšćˆ‡ć‰ČéąäžŠçš„çœźäžș1ïŒŒćˆ‡ć‰Čéąäž‹çš„çœźäœ-1ïŒŒèż›èĄŒć€šæŹĄćˆ‡ć‰Čäč‹ćŽïŒŒäž€äžȘ掟ç©ș闎盞䌌的物䜓生成的äșŒèż›ćˆ¶çŒ–ç ć°±æŻ”èŸƒç›žäŒŒïŒŒè€ŒćŽŸç©ș闎䞍盞䌌的物䜓生成的Hash码äčŸć°±äžç›žäŒŒă€‚有äșșćœšæ–‡ç« äž­ć†æćŠéšćłæ˜ ć°„hash时èźČćˆ°ç±»äŒŒäșŽäž€ç§Boostæ–čæł•ïŒŒćłć€šäžȘHashć˜æˆäž€äžȘæˆ‘ä»ŹæƒłèŠçš„ćŒșHashæ–čæł•
  9. äžŽæ•°æźç›žć…łçš„Hashçź—æł•ć€§ć€šæ˜ŻæŠŠçŽ°æœ‰çš„æœș晹歩äč æ–čæł•èż›èĄŒć„—ç”šïŒŒäŸ‹ćŠ‚NIPS2008äžŠçš„èż™çŻ‡æ–‡ç« ïŒŒéŠ–ć…ˆæ čæźæˆ‘ä»ŹèŠäŒ˜ćŒ–çš„ç›źçš„æž„é€ Hashć‡œæ•°ïŒŒć‡èźŸyæ˜Żæ˜ ć°„ćŽçš„Hash码Sæ˜Żç›žäŒŒćșŠçŸ©é˜”ïŒŒé‚Łäčˆćœ“xy盞䌌ćșŠSæŻ”èŸƒć€§çš„æ—¶ć€™ïŒŒyi,yjćˆ™éœ€èŠæŻ”èŸƒć°ïŒŒćœ“ç›žäŒŒćșŠSæŻ”èŸƒć°çš„æ—¶ć€™ćˆ™yi,yjćŻä»„ć€§äž€ç‚čă€‚ćŠ‚æžœç›ŽæŽ„äŒ˜ćŒ–èż™äžȘç›źæ ‡ć‡œæ•°çš„æ—¶ć€™ć­˜ćœšäž€äžȘćčłć‡Ąè§ŁïŒŒć°±æ˜Żyć€Œéƒœäž€æ ·ïŒŒć› è€Œéœ€èŠćŠ ć…„äž‹éąäž‰äžȘæĄä»¶ćšçșŠæŸïŒŒçŹŹäž€äžȘæ˜ŻèŻŽyæ˜ŻäșŒć€ŒćŒ–çš„ïŒŒçŹŹäșŒäžȘæ˜ŻèŻŽçŒ–ç äž­çš„-1,1ćˆ†ćžƒć‡ćŒ€ïŒŒçŹŹäž‰äžȘćźžèŻŽäžćŒç»Žäč‹é—Žäș’äžç›žć…łă€‚ćœšèż™æ ·çš„çșŠæŸæĄä»¶äž‹ćŸˆéšŸæ±‚è§ŁïŒŒæˆ‘ä»Źæ”ŸćźœäșŒć€ŒćŒ–æĄä»¶ïŒŒć°±èż™äžȘé—źéą˜ć°±é€€ćŒ–æˆè°±èšç±»çš„é—źéą˜ïŒŒè°±èšç±»ç›žć…łçš„ç ”ç©¶ćŸˆć€šïŒŒæœ€ćžžç”šçš„è§Łæł•ć°±æ˜Żæ‹‰æ™źæ‹‰æ–Żç‰čćŸæ˜ ć°„çš„æ–čæł•èż›èĄŒæ±‚è§ŁïŒŒćŸ—ćˆ°ćźžć€ŒćŒ–çš„YïŒŒç„¶ćŽèźŸćźšé˜ˆć€Œèż›èĄŒäșŒć€ŒćŒ–ćŸ—ćˆ°Hashç ă€‚ćœšèż™é‡Œæ‰Żäž€äžȘéą˜ć€–èŻïŒŒć› äžșćœšæ”èĄŒć­Šäč äž­äŒšæ¶‰ćŠćˆ°è°±èšç±»ć’Œæ‹‰æ™źæ‹‰æ–Żç‰čćŸæ˜ ć°„ïŒŒæœ‰æ—¶èźș文䞭äčŸäŒšæćˆ°â€œć”Œć…„ćœšé«˜ç»Žç©șé—Žäž­çš„äœŽç»Žæ”ćœąâ€ïŒŒćż…èŻŽćŠ‚ïŒŒć€§ćœ°ć…¶ćźžć°±æ˜Żć”Œć…„ćˆ°äž‰ç»Žç©ș问侭的äșŒç»Žæ”ćž‹ïŒŒä»Žć±€éƒšæ„èźČïŒŒć€§ćœ°æ˜ŻäžȘäșŒç»ŽćčłéąïŒŒè€Œèżœè§‚æ„çœ‹ć€§ćœ°æ˜ŻäžȘäž‰ç»Žçƒćž‹ïŒŒè€Œć…¶ćźžäșŒç»Žćčłéąçš„ç‰čæ€§æ›Žé‡èŠă€‚ćœšćˆ©ç”šè°±èšç±»èż›èĄŒé™ç»Žçš„èż‡çš‹äž­æˆ‘ä»Źæž„ć»șçš„æ˜Żäž€äžȘćžŠæœ‰ç›žäŒŒćșŠçš„ć…łè”çŸ©é˜”ïŒŒè€Œæ‹‰æ™źæ‹‰æ–Żç‰čćŸæ˜ ć°„æŠ“äœäș†äž»èŠç‰čćŸïŒŒé‚Łäčˆæ˜ ć°„ćˆ°äœŽç»Žäč‹ćŽäżç•™äș†ćŽŸç©șé—Žäž­çš„ć†…ćœšæ€§èŽšă€‚æŻ”ćŠ‚èŻŽćłäž‹è§’èż™ćč…ć›ŸïŒŒćŠ‚æžœé‡‡ç”šæ‹‰æ™źæ‹‰æ–Żç‰čćŸæ˜ ć°„ć’Œé‡‡ç”šK-meansèšç±»çš„ç»“æžœè‚Żćźšäžäž€æ ·ïŒŒæŻ”ćŠ‚èŻŽèż™ćč…ć›Ÿäž­çš„äž€äžȘç‚č的距犻ćŸșäșŽæŹ§ćŒç©ș问的K-meansæ–čæł•ć’Œ æ”èĄŒć­Šäč æ–čæł•çš„ç†è§Łć°±äžäž€æ ·ă€‚ć„œæˆ‘ä»Źć›žćˆ°æ­Łäœ“ă€‚
  10. ćˆ°èż™é‡ŒćŻèƒœäŒšæœ‰äž€äș›ç–‘é—źïŒŒäž€äžȘæ˜Żæˆ‘ä»Źèƒœäžèƒœç›ŽæŽ„é€šèż‡äșŒć€ŒćŒ–䞀äžȘäœŽç»Žçš„ćźžć€Œć‘é‡ćŸ—ćˆ°HashCode 揩侀äžȘïŒŒæ—ąç„¶DeepLearningæ–čæł•äž­æŻć±‚çš„ç‰čćŸć€§ć€šæ˜ŻäșŒć€ŒćŒ–èĄšç€șçš„ïŒŒé‚Łäčˆèƒœäžèƒœæ‹żć…¶äž­æŸć±‚çš„ç‰čćŸćšHashCode
  11. æČĄæœ‰é—źéą˜ïŒŒæˆ‘ä»ŹćŻä»„çœ‹äž€äž‹07ćčŽSemanticHashing的文章䜜者 萚拉蔫䞁èŻș怫 撌Hintonèż™äž€äžȘäșș性柶ćș”èŻ„æŻ”èŸƒç†Ÿæ‚‰ïŒŒ06ćčŽćžˆćŸ’äż©ćœšScienceäž€è”·ć‘èĄšçš„èźș文掀蔷DeepLearning新æ”ȘæœźïŒŒè€Œćœšä»–ä»ŹSIGIR2007ćčŽçš„Semantic Hashingæ–‡ç« äž­ć°±æ˜Żç”šçš„RBMèź­ç»ƒæ•°æźïŒŒæ‹żć…¶äž­äșŒć€ŒćŒ–çš„ç‰čćŸć±‚èż›èĄŒHashCode而䞔他的BaseLineèŻ•éȘŒć°±æ˜Żç”šLSIèż›èĄŒé™ç»Žćˆ†è§Łäč‹ćŽèźŸćźšé˜ˆć€ŒćŸ—ćˆ°äșŒć€ŒćŒ–ć‘é‡. äžèż‡ć°±æ˜ŻèŻŽïŒŒèż™ç§ç›ŽæŽ„äșŒć€ŒćŒ–çš„æ–čæł•äž€èˆŹćŸ—ćˆ°çš„æ•ˆæžœäžæ˜ŻćŸˆć„œă€‚èšæ‹‰è”«äžèŻșć€«ćšćŁ«èźșæ–‡çš„çŹŹäșŒç« ć°±æ˜ŻSemantic Hashing䜆从07ćčŽè‡łä»Šç”šDeepLearningæ„ćšHashingçš„æ–‡ç« æœ‰ïŒŒäœ†ćŸˆć°‘ïŒŒäžèż‡ć€§ćź¶ćœšèźČ戰Semantic Hashingèż™äžȘèŻć’Œç”šLSI恚BaseLineèŻ•éȘŒçš„æ—¶ć€™èż˜æ˜ŻéƒœäŒšćŸˆç€ŒèČŒçš„ćŒ•ç”šäž€äž‹ä»–ä»Źèż™çŻ‡èźșæ–‡ă€‚
  12. 那äčˆćœš13ćčŽèż‡ćŽ»çš„䞀ćčŽïŒŒć€§ćź¶ćœšHashingæ–č靱恚äș†äș›ä»€äčˆïŒŸæˆ‘çź€ć•è°ƒç ”äș†äž€äž‹ïŒŒćœšæˆ‘调研的16äžȘć›œé™…äŒšèźźäž­èźŸèźĄHashingé—źéą˜çš„ć…±æœ‰30çŻ‡ïŒŒäž‹éąæˆ‘æŒ‘ć‡ș侀äș›ćœšèż™é‡Œä»‹ç»ïŒŒäž»èŠćȘä»‹ç»ä»–ä»Źçš„ideaïŒŒćŽ»ćčŽć‘èĄšHashingæœ€ć€šçš„æ˜ŻIJCAI, CVPR撌CIKMéƒœæ˜Ż5çŻ‡ïŒŒè€ŒVLDB侀篇æČĄæœ‰ïŒŒäœ†ćźžé™…äžŠHash最早的èźșæ–‡æ˜Żć‡șè‡ȘVLDB的。
  13. èż™æ—¶äž€çŻ‡ICMLäžŠçš„æ–‡ç« ïŒŒæ˜Żé€šèż‡ćˆ—ç”Ÿæˆæł•ć­Šäč Hashć‡œæ•°ïŒŒæˆ‘æŠŠä»–çš„äœœè€…äżĄæŻć’Œç»„ç»‡ć•äœäčŸćˆ—ćœšèż™é‡ŒïŒŒć€§ćź¶ćŻèƒœäŒšćˆ°çœ‹äž€äș›æŻ”èŸƒç†Ÿæ‚‰çš„ćŒèĄŒïŒŒäčŸćŻä»„äș†è§Łäž€äž‹æœ‰ć“Șäș›äșșćœšćšèż™ä»¶äș‹ïŒŒćŒæ—¶ć°œćŻèƒœçš„æŠŠćźƒçš„ć·„䜜ćŸș础äčŸćˆ—ćœšèż™é‡ŒïŒŒèƒœçœ‹ć‡șä»–æ˜ŻćŸșäșŽè°çš„ć·„äœœç»§ç»­ć‰èż›ă€‚æˆ‘ä»Źé‡ç‚čèż˜æ˜Żçœ‹ä»–çš„Idea, èż™çŻ‡æ–‡ç« çš„Motivationć°±æ˜ŻèŻŽïŒŒćœšćŸșäșŽć†…ćźčçš„ć›Ÿç‰‡æŁ€çŽąäž­ïŒŒæœé›†ç”šæˆ·ćéŠˆäżĄæŻçš„æ—¶ć€™ïŒŒèź©äž€äžȘç”šæˆ·æ„æ„ćˆ€æ–­äž€ćŒ ć›Ÿç‰‡XćŻčæŻ”äșŽX-ïŒŒæ˜ŻćŠć’ŒX+æ›Žç›žäŒŒă€‚èż™æ ·çš„ä»»ćŠĄé€šćžžèŠæŻ”æŻćŒ ć›Ÿç‰‡èŽŽæ ‡ç­Ÿćźčæ˜“çš„ć€šă€‚ć› è€Œä»–ćœšç›źæ ‡ć‡œæ•°é‡Œéąć°±æž„é€ äș†äž€äžȘçșŠæŸéĄčx撌x-çš„è·çŠ»èŠæŻ”x撌x+çš„è·çŠ»ć€§ïŒŒé€šèż‡ćˆ—ç”ŸæˆäŒ˜ćŒ–çź—æł•æ„æ±‚è§Łèż™äžȘé—źéą˜ă€‚çź—æł•ç»†èŠ‚æˆ‘ä»ŹéƒœäžćšèźšèźșćȘć…łćżƒä»–ä»Źćšèż™ä»¶äș‹çš„Motivation撌Ideaă€‚ć·Šäž‹è§’æ˜Żä»–çš„èŻ•éȘŒç»“æžœïŒŒçŹŹäž€ć‰Żæ˜ŻèŸ“ć…„äž€ć°ç””è„‘çš„æŁ€çŽąç»“æžœïŒŒäž‹éąæ˜ŻMNISTæ•°æźé›†äžŠçš„æ•°æźèŻ†ćˆ«æ•ˆæžœ
  14. èż™çŻ‡ICMLçš„æ–‡ç« æ˜ŻçŸŽć›œé©Źé‡Œć…°ć€§ć­Šçš„ć·„äœœïŒŒä»–çš„Motivationæ˜ŻçŽ°ćœšèŽ·ć–äżĄæŻäž€èˆŹéƒœæ˜Żæ„è‡Ș怚äžȘæž é“ïŒŒäŸ‹ćŠ‚ć›Ÿç‰‡ć’ŒćŻčćș”çš„æ–‡ć­—æèż°ă€‚ćŒæ—¶ä»Žèż™äž€äžȘéą†ćŸŸèż›èĄŒhashçŒ–ç ă€‚æœ€ç»ˆèƒœć€ŸćźžçŽ°æŠŠæ–‡ć­—ć’Œć›Ÿç‰‡æ˜ ć°„ćˆ°ćŒäž€äžȘHammingç©șé—Žäž­èż›èĄŒäżĄæŻæŁ€çŽąă€‚äŸ‹ćŠ‚ćłäž‹è§’æ˜Żä»–çš„èŻ•éȘŒæ ·äŸ‹ïŒŒæœ€äž‹éąäž€èĄŒïŒŒèŸ“ć…„Plane flying on the air,果真蟓ć‡ș侀äș›éŁžæœșćœšć€©äžŠéŁžçš„ç…§ç‰‡ă€‚ćŒæ ·ïŒŒæœ€äžŠéąèŸ“ć…„Laptop placed on the table, 蟓ć‡șçš„ć›Ÿç‰‡ç»“æžœäž­ïŒŒäžćȘ有laptopïŒŒèż˜æœ‰television
  15. èż™çŻ‡ICMLçš„æ–‡ç« æ˜ŻçŸŽć›œæ™źæžĄć€§ć­Šçš„ć·„äœœïŒŒä»–ä»ŹćźžéȘŒćź€çš„ć€§è€æżæ˜ŻäžȘ捎äșșSi Luo,äč‹ć‰ćœšæž…ćŽèŻ»ćźŒæœŹç§‘ć’ŒçĄ•ćŁ«äč‹ćŽïŒŒè·‘ćˆ°CMUèŻ»äș†çĄ•ćŁ«ć’ŒćšćŁ«ïŒŒç„¶ćŽćˆ°æ™źæžĄć€§ć­Šä»»æ•™ïŒŒäž“æłšäșŽäżĄæŻæŁ€çŽąïŒŒæœș晹歩äč ć’Œè‡Șç„¶èŻ­èš€ć€„ç†ă€‚ç„¶ćŽä»–çš„ć€§éƒšćˆ†ć­Šç”Ÿéƒœæ˜Żä»Žæž…ćŽć€§ć­Šèż‡ćŽ»æ·±é€ çš„ă€‚ä»–ä»ŹćźžéȘŒćź€çš„IRæ–č靱恚äș†ćŸˆć€šæŻ”蟃有意äč‰çš„ć·„äœœă€‚èż™çŻ‡æ–‡ç« çš„Motivation杄è‡Ș䞀æ–čéąïŒŒ1).èŻŽçŽ°æœ‰çš„Hashæ–čæł•æČĄæœ‰è€ƒè™‘æ ‡ç­ŸäżĄæŻïŒŒè€Œćźžé™…äžŠäč‹ć‰çš„éƒšćˆ†Hashingć·„äœœèż˜æ˜Żè€ƒè™‘äș†HashäżĄæŻçš„ïŒŒæ‰€ä»„ä»–ćŠ äș†fullyèż™æ ·çš„äżźé„°ïŒ›2). 珏äșŒæ–čéąèŻŽä»„ć‰çš„Hashingć·„äœœéƒœæ˜Żé€šèż‡æ–‡æœŹçš„ćŽŸć§‹ç‰čćŸæ„äżć­˜æ–‡æœŹé—Žçš„ç›žäŒŒäżĄæŻçš„ïŒŒèż™äžȘäžèƒœć……ćˆ†ćæ˜ æ–‡æœŹé—Žçš„èŻ­äč‰ć…łçł»ïŒŒć› è€Œä»–ćŒ•ć…„äș†TopicModeling。掻ćčŽæˆ‘ć†™æ–‡ç« çš„æ—¶ć€™äčŸćŻčèż™æ–čéąć·„äœœèż›èĄŒäș†è°ƒç ”ïŒŒçĄźćźžèż˜æČĄæœ‰äșș朹Hashæ–čæł•äž­ćŒ•ć…„TopicModleć·„äœœçš„ïŒŒäčŸæ˜Żćˆç”Ÿç‰›çŠŠäžæ€•è™Žçš„氱憙äș†äž€çŻ‡ćŸșäșŽTopicç‰čćŸçš„ćż«é€ŸæŁ€çŽąæ–čæł•ă€‚æ‰€ä»„ä»–ä»Źèż™çŻ‡æ–‡ç« äž­äčŸæ˜ŻæŒ‡æ˜ŽïŒŒAs far as we knowïŒŒä»–ä»Źæ˜ŻçŹŹäž€äžȘćŒ•ć…„TopicModel戰Hashäž­çš„ć·„äœœă€‚ćŒć­äž­ïŒŒthetaæ˜Żäž»éą˜ç‰čćŸïŒŒTæ˜Żæ ‡ç­ŸäżĄæŻïŒŒUæ˜Żé’ˆćŻč标筟ć»șæšĄçš„äž€äžȘ隐揘量。çșŠæŸæĄä»¶æ˜ŻćŻčHash码的çșŠæŸă€‚ćłäž‹è§’æ˜Żä»–ä»Źćœš20NewsGroups撌WebKBäžŠçš„èŻ•éȘŒç»“æžœïŒŒć…¶ćźžæ•ˆæžœæć‡ćč¶äžæ˜ŻćŸˆæ˜Žæ˜Ÿă€‚æŻ•ç«Ÿä»–çš„BaseLine陀SSHïŒŒć…¶ä»–çš„éƒœæČĄæœ‰äœżç”šæ ‡ç­ŸäżĄæŻ
  16. ć·ŠäžŠć›Ÿæ˜Żæˆ‘ä»Źé€šèż‡ćŠäž€ç§æ–čćŒćŒ•ć…„TopicFeature撌TagäżĄæŻćœš20NewsGroups侊恚的ćŸșäșŽHashingçš„æ–‡æœŹæŁ€çŽąç»“æžœïŒŒä»ŽćŹć›ž-çČŸćșŠæ›ČçșżäžŠèƒœ,ćŻä»„çœ‹ć‡șćŠ ć…„æ ‡ç­ŸäżĄæŻćŽçš„ç»“æžœæ•ˆæžœćș”èŻ„æ˜Żèƒœć€Ÿæ”čć–„ćŸˆć€šçš„ă€‚äžèż‡SIGIRèż™çŻ‡æ–‡ç« äžșäș†èŻæ˜Žè‡Șć·±çź—æł•çš„ćŻčTagäżĄæŻçš„éČæŁ’æ€§ïŒŒćœšèŻ•éȘŒæ•°æźäžŠéšćłćˆ æŽ‰äș†éƒšćˆ†TagäżĄæŻă€‚
  17. èż™çŻ‡IJCAIçš„æ–‡ç« æ˜Żæ”™æ±Ÿć€§ć­Šçš„ć·„äœœïŒŒèż™é‡Œæœ‰äž€äžȘć€§ç‰›ïŒŒäœ•æ™“éŁžć’Œè”Ąç™»ïŒŒäœ•æ™“éŁž00ćčŽä»Žæ”™ć€§æŻ•äžšćŽćˆ°èŠćŠ ć“„ć€§ć­Šæ·±é€ ćšćŁ«ïŒŒ05ćčŽæŻ•äžšćŽèż›ć…„Yahooç ”ç©¶é™ąïŒŒäž€ćčŽćŽ07ćčŽäčŸć°±æ˜Ż29ćČæ—¶èą«æ”™ć€§é«˜è–Șè˜ç”šćŠ ć…„èźĄçź—æœșć­Šé™ąçš„ć›œćź¶é‡ç‚č漞éȘŒćź€ïŒŒ1ćčŽć·Šćłæ‹żćˆ°æ­Łæ•™æŽˆă€‚ćœšć›œć†…æœș晹歩äč ć’ŒèźĄçź—æœș视觉äčŸæ˜ŻćŒ•è”·äș†ć°ć°ćœ°éœ‡ïŒŒäž€èˆŹä»Žć­ŠæœŻç•ŒèœŹèĄŒćˆ°ć·„äžšç•Œć€§ć€šæ˜Żç‰©èŽšèŻ±æƒ‘ïŒŒè€Œä»Žć·„äžšç•ŒèœŹćˆ°ć­ŠæœŻç•Œć€§ć€šæ˜Żé€ èŻŁćŸˆæ·±çš„ă€‚ä»–çŽ°ćœšçš„èźșæ–‡ćŒ•ç”šçŽ‡ć·Č超1WæŹĄă€‚è”Ąç™»ćˆ™æ˜Ż08ćčŽä»ŽçŸŽć›œćšćŁ«æŻ•äžšćŽćŠ ć…„äș†æ”™ć€§èźĄçź—æœșçł»ćšć‰Żæ•™æŽˆïŒŒè·Ÿéšäœ•æ™“éŁžă€‚éĄșé“éą˜ć€–èŻïŒŒäčŸ8gäž€äž‹èż™äž€äœç‰›äșșïŒŒè”Ąç™»ćœšçŸŽć›œèŻ»äčŠæ—¶ä»ŽćžˆéŸ©ćź¶ç‚œïŒŒéŸ©æ•™æŽˆæ˜ŻKDDäŒšèźźçš„ć‘è”·è€…ïŒŒäœ†äčŸæœ‰äžȘć€–ć·ć«ćŽäșșç•Œçš„çŒæ°Žć€©çŽ‹ă€‚è€ŒäŒ èš€äœ•æ™“éŁžćœšèŻ»ćšæœŸé—Žé€šèż‡äž€äžȘidea灌氎èźș文30çŻ‡ă€‚ćœ“ç„¶èż™äčŸæ˜ŻćŒèĄŒäč‹é—Žæ–‡äșșç›žèœ»ïŒŒæŻ•ç«Ÿèż™äž€äžȘ牛äșșäž€è”·æ‹żćˆ°èż‡12ćčŽAAAI的BestPaperæ˜Żćšæ‘˜èŠçš„ć’Œ10ćčŽCAM-MM的BestPaperæ˜ŻćšćŸșäșŽç€Ÿäș€ćȘ’äœ“çš„éŸłäčæŽšèă€‚ æˆ‘ä»Źć›žæ„çœ‹IJCAIçš„èż™çŻ‡æ–‡ç« ïŒŒä»–çš„Motivationæ˜ŻèŻŽïŒŒäŒ ç»Ÿæ–čæł•äž­è§Łć†łèż‘é‚»æœçŽąé—źéą˜ïŒŒäșșä»Źé€šćžžè‡ŽćŠ›äșŽæž„ć»șæœ‰æ•ˆçš„æ•°æźç»“æž„ïŒŒäŸ‹ćŠ‚ć±‚æŹĄçš„K-means Tree,æˆ–è€…äœżç”šHashingæ–čæł• æ„ćŠ é€ŸæŸ„èŻąèż‡çš‹ă€‚ćœšèż™çŻ‡æ–‡ç« äž­ïŒŒä»–æć‡șäž€ç§é€Œèż‘çš„èż‘é‚»æœçŽąæ–čæł•ïŒŒć°±æ˜Żæ‰ćˆćŸșäșŽTreeçš„æ•°æźç»“æž„ć’ŒćŸșäșŽHammingèźĄçź—çš„Hashæ–čæł•ă€‚æƒłæł•æŻ”èŸƒæž…æ™°ïŒŒè€Œäž”äžéœ€èŠéąć€–ä»»äœ•çš„ć…ŹćŒæŽšćŻŒïŒŒæ‰€ä»„ä»–èż™çŻ‡æ–‡ç« æČĄæœ‰ä»»äœ•æ•°ć­Šć…ŹćŒïŒŒćȘæ˜Żç»™ć‡șäž€ć‰Żć›ŸäŸ‹ïŒŒç„¶ćŽć°±ćŒ€ć§‹èŻ•éȘŒă€‚ć·Šć›Ÿæ˜ŻçŠ»çșżæž„造的䞀äžȘK-means TreeïŒŒćłć›Ÿæ˜Żqueryèż‡çš‹ïŒŒćŠ ć…„Query节ç‚čć·Čç»ćˆ°çŹŹić±‚ïŒŒStep(a)æ˜Żé€šèż‡Hashæ–čæł•æ‰Ÿćˆ°æœ€ç›žäŒŒçš„4äžȘ䞭濃ç‚čïŒŒç„¶ćŽStep(b)é€šèż‡çČŸçĄźçš„Rankçź—æł•ćŸ—ćˆ°æœ€äžș盞䌌的䞀äžȘ䞭濃ç‚čïŒŒç„¶ćŽæŻäžȘç±»ć†…ćˆæœ‰äž‰äžȘć­ç±»ïŒŒQueryć°±èż›ć…„äș†i+1ć±‚ă€‚æƒłæł•æŻ”èŸƒçź€ć•ïŒŒäžèż‡äčŸæ˜Żäž€ç§æŻ”èŸƒć„œçš„ćż«é€ŸæŁ€çŽąæĄ†æž¶ïŒŒæŻ•ç«Ÿćœšć…šæ•°æźé›†äžŠçš„HashæŁ€çŽąç»“æžœèż˜æ˜Żć€Șć€šïŒŒć› è€ŒćȘè·Ÿèšç±»çš„äž­ćżƒèŠ‚ç‚čćšć±‚æŹĄćŒ–ćŻčæŻ”ă€‚
  18. æ—ąç„¶èźČ戰Hash撌K-MeansïŒŒæˆ‘ä»Źæ„çœ‹ćŠäž€çŻ‡ćˆ©ç”šHash撌K-Meansçš„ć·„äœœïŒŒæ˜ŻćŸźèœŻäșšæŽČç ”ç©¶é™ąçš„äœ•æșæ˜Žçš„ć·„äœœïŒŒäœ•æșæ˜ŽćœšCV界äčŸçź—æ˜Żç‰›äșșäș†ă€‚09ćčŽCVPR的BestPaperïŒŒć°±æ˜Żć†ŻćȘ›ćŽ»ćčŽèźČèż‡äž€çŻ‡æ–‡ç« æ˜Żć…łäșŽć›ŸćƒćŽ»é›Ÿçš„ć·„äœœă€‚æˆ‘ä»Źçœ‹äž€äž‹ä»–çš„MotivationèźČ戰Hamming-basedæ–čæł•ć’ŒLookup-basedæ–čæł•ćœšæŁ€çŽąéą†ćŸŸć—ćˆ°ćčżæł›ć…łæłšïŒŒćœšäžćŒéą†ćŸŸä»–ä»Źæœ‰ć„è‡Ș的䌘ç‚čïŒŒćŠ‚ćœšćŒæ ·ç»Žæ•°çš„çŒ–ç äžŠLookup-basedæ–čæł•èŠæ›ŽçČŸçĄźïŒŒè€ŒHamming-basedæ–čæł•ćˆ™æŸ„èŻąæ›Žćż«ă€‚ćŠ‚ć›Ÿäž­ć±•ç€ș的Hamming-based的Hashæ–čæł•æ˜ŻæŠŠç‰©äœ“æ˜ ć°„ćˆ°äșŒèż›ćˆ¶çš„Hammingç©ș闎䞭而Lookup-based的K-meansæ–čæł•æ˜Żæ‰Ÿćˆ°ć…·æœ‰ä»ŁèĄšæ€§çš„ç±»äž­ćżƒç‚č。那äčˆèƒœäžèƒœćŒæ—¶ć­Šäč ïŒŒæŠŠhashæ˜ ć°„èż‡çš‹ć’Œèšç±»èż‡çš‹ćŒæ—¶èż›èĄŒïŒŒä»ŽèŠäŒ˜ćŒ–çš„ç›źæ ‡ć‡œæ•°äčŸèƒœçœ‹ć‡șæ„ć°±æ˜ŻèŠK-meansçš„èšç±»äž­ćżƒç‚čèż‘çš„èŻïŒŒHashæ˜ ć°„ćŽçš„Hamming距犻äčŸèŠèż‘ïŒŒćäč‹äșŠç„¶ă€‚掻ćčŽćˆ˜éčçš‹ćœšä»‹ç»èż™çŻ‡æ–‡ç« çš„æ—¶ć€™ïŒŒæćˆ°äž€äžȘæŻ”èŸƒæœ‰æ„æ€çš„äș‹æƒ…ïŒŒć°±æ˜Żäœœè€…ćœšè‡Ș深的Paper侭柋äș†äž€äžȘćœ©è›‹ïŒŒçź—æł•çš„çź€ç§°KMHć’Œäœœè€…ćć­—çš„çź€ç§°äž€è‡Žă€‚
  19. èż™çŻ‡ICCV的文章äčŸæ˜Żć‡șè‡Șæ”™ć€§çš„è”Ąç™»ćźžéȘŒćź€ïŒŒćŒæ—¶èż˜æœ‰ç™ŸćșŠç ”ç©¶é™ąć’Œäž­ç§‘é™ąè„żćź‰ćˆ†é™ąçš„ć‚äžŽă€‚èż˜èź°ç€æˆ‘ä»Źäč‹é—ŽèźČèż‡èŻŽćŸșäșŽéšćłæ˜ ć°„çš„Hashæ–čæł•ć°±æ˜Żćœšé«˜ç»Žç©șé—Žäž­èż›èĄŒè¶…ćčłéąćˆ‡ć‰Č。ćŸșäșŽæ­€ïŒŒä»–仏的Motivation有䞀ç‚č1ćŠ‚ć·Šć›ŸïŒšćŸșäșŽéšćłæ˜ ć°„çš„Hashæ–čæł•ïŒŒćŠ‚æžœè¶…ćčłéąaæ˜Żç©żèż‡äž€ç±»äž­é—Žçš„çš€ç–ćŒșćŸŸïŒŒé‚ŁäčˆäžćŒç±»ćˆ™ć„è‡Ș量挖戰搌侀äžȘæĄ¶äž­ă€‚ćŠ‚æžœè¶…ćčłéąbç©żèż‡æ•°æźçš„ćŻ†é›†ćŒșçš„èŻćˆ™æ•°æźäŒšèą«é‡ćŒ–ćˆ°äžćŒçš„æĄ¶äž­ă€‚Motivation2æ˜ŻèŻŽïŒŒè¶…ćčłéąa撌béƒœæŠŠæ•°æźćˆ†ćŒ€ïŒŒäœ†æ˜Żèż™äž€äžȘćčłéąäž€è”·äœœç”šæ—¶ïŒŒćŠ‚æžœæ˜Żcçš„æƒ…ć†”ïŒŒćˆ™äčŸäžæ˜Żäž€äžȘćŸˆć„œçš„æ˜ ć°„ćčłéąïŒŒćŠ‚æžœç»„ćˆæˆdçš„èŻïŒŒćˆ™æ˜Żæœ€ć„œçš„ïŒŒäž€äžȘ超ćčłéąć°±ćŻä»„把4ç»„æ•°æźćŒșćˆ†ćŒ€ïŒŒè€Œäž”ćŒäž€ç»„æ•°æźæ˜ ć°„ćˆ°ćŒäž€äžȘæĄ¶äž­ă€‚
  20. CVPRçš„èż™çŻ‡æ–‡ç« æ˜Żç”±ćŒ—èˆȘïŒŒć“„äŒŠæŻ”äșšć€§ć­ŠćŠFacebookäž€è”·ć‚äžŽçš„ć·„äœœïŒŒä»–ä»Źçš„Motivationæ˜ŻèŻŽè™œç„¶ç›źć‰ćœšć€§æ•°æźçš„èż‘é‚»æœçŽąé—źéą˜äžŠHashing的ćș”甚ć·Čç»ćŸˆæˆćŠŸïŒŒäœ†æ˜Żä»ç„¶æœ‰äž€äș›é—źéą˜éœ€èŠèźšèźșïŒŒć°±æ˜Żć€šç‰čćŸçš„é€‰æ‹©é—źéą˜ïŒŒć€šHashingçź—æł•çš„é€‰æ‹©ïŒŒä»„ćŠć€šć‚æ•°èźŸćźšçš„é€‰æ‹©ă€‚æœŹæ–‡ç»™ć‡șäș†äž€äžȘæĄ†æž¶ïŒŒé€šèż‡é’ˆćŻč搌侀äžȘæ•°æźé›†é‡‡ç”šäžćŒç‰čćŸïŒŒäžćŒHashingçź—æł•ïŒŒäžćŒć‚æ•°èźŸćźšäž‹ç”Ÿæˆć„ç§äžćŒçš„äșŒć€Œć‘é‡ïŒŒæž„æˆć›ŸïŒŒç„¶ćŽć†ä»Žć›Ÿäž­é€‰ć‡ș最䌘节ç‚č。
  21. ICCVèż™çŻ‡æ˜ŻæŸłć€§ćˆ©äșšçš„é˜żćŸ·é›·ćŸ—ć€§ć­žçš„ć·„äœœïŒŒä»–ä»Źçš„Motivationæ˜ŻèŻŽç›źć‰ć€§ć€šæ•°Hashæ–čæł•éƒœæ˜Żé’ˆćŻčæ•°æźé›†çš„Hashé™ç»ŽçŒ–ç èż‡çš‹ćŠHashçŒ–ç éą„æ”‹ć‡œæ•°çš„ć­Šäč èż‡çš‹æ•Žćˆćœšäž€è”·æ„ć­Šäč ïŒŒèż™ç§çŽ§ćŻ†è€Šćˆäž€æ–čéąé™ćˆ¶äș†ç”æŽ»æ€§ïŒŒćŠäž€æ–čéąćŻŒè‡ŽäŒ˜ćŒ–é—źéą˜ć˜çš„ć€æ‚ïŒŒéšŸä»„æ±‚è§Łă€‚ä»–ä»Źæć‡șäž€ç§æĄ†æž¶ïŒŒæŠŠHashé—źéą˜æ‹†è§Łæˆäž€äžȘé˜¶æź”ćźŒæˆïŒŒçŹŹäž€äžȘé˜¶æź”æ˜Żèż›èĄŒé’ˆćŻčçŽ°æœ‰çš„æ•°æźé›†èż›èĄŒhash码歊äč ïŒŒçŹŹäșŒäžȘé˜¶æź”æ˜ŻćŸșäșŽäč‹ć‰çš„Hash码歊äč Hashć‡œæ•°ă€‚ćŠ‚æžœćŻčMotivation䞍ć€Șæž…æ„šçš„ïŒŒæˆ‘ä»ŹćŻä»„äž‹éąèż™äžȘć›ŸäŸ‹ïŒŒèż™äžȘć›ŸäŸ‹æ˜Żèż™çŻ‡æ–‡ç« çš„äž»èŠć‚è€ƒć·„äœœSIGIR2010的Self-Taught HashingïŒŒćźƒć°±æ˜Żäž€äžȘ慾枋 äșŒé˜¶æź”Hashć­Šäč æ–čæł•ïŒŒć‡șè‡ȘäșŽæ™źæžĄć€§ć­Šçš„Si Luo漞éȘŒćź€ïŒŒèż™çŻ‡æ–‡ç« çš„çŹŹäž‰äœœè€…æ˜Żæ”™ć€§çš„è”Ąç™»ïŒŒćŻèƒœæ˜Żäș€æ”ć­Šäč é˜¶æź”äž€è”·ćźŒæˆçš„ć·„äœœă€‚æˆ‘ä»Źçœ‹èż™äžȘć›ŸïŒŒéŠ–ć…ˆç»™äž€ć †æ–‡æœŹé›†ç„¶ćŽé€šèż‡äž€ç§æ— ç›‘çŁçš„é™ç»Žæ–čæł•ćŸ—ćˆ°äșŒèż›ćˆ¶çš„Hashç ïŒŒèż™æ˜ŻçŹŹäž€äžȘé˜¶æź”ă€‚ç„¶ćŽæ čæźć·Čç»ć­Šćˆ°çš„Hash码䜜äžșäșŒć€Œæ ‡ç­Ÿćˆ©ç”šç›‘督歩äč æ–čæł•ć­Šäč äž€äžȘHashć‡œæ•°ă€‚è€Œèż™äž€äžȘé˜¶æź”éƒœć±žäșŽçŠ»çșżć­Šäč é˜¶æź”而QueryæŸ„èŻąć±žäșŽćœšçșżé˜¶æź”。慶柞STHæœŹèș«ć°±æ˜Żäž€äžȘäșŒé˜¶æź”æĄ†æž¶äș†ïŒŒICCVçš„èż™çŻ‡æ–‡ç« ćŸșæœŹć°±æ˜ŻćŸșäșŽæ­€ć·„䜜提ć‡șæ€»ç»“æ€§çš„äž€é˜¶æź”Hashć­Šäč æĄ†æž¶
  22. èż™çŻ‡IJCAIæ˜Żäž­ć±±ć€§ć­Šçš„ć·„äœœïŒŒä»–ä»Źçš„ć·„äœœæ˜ŻćŸșäșŽDMKD2012ćčŽäžŠäž€çŻ‡ćŸșäșŽäž»ćŠšć­Šäč Hash的文章(DMKDæ˜ŻæŁ€çŽąç±»é‡Œéąçš„Bçș§æœŸćˆŠ)。他仏的MotivationæŻ”èŸƒćźžç”šćŒ–ïŒŒć°±æ˜ŻèŻŽçŽ°æœ‰çš„ćŸșäșŽHash的æ–čæł•ć·Čç»èŽ·ćŸ—äș†æŻ”èŸƒć„œçš„æ•ˆæžœïŒŒäœ†æ˜Żä»–ä»Źć€§ć€šæ˜Żèą«ćŠšHashingć­Šäč ïŒŒäž”ć‡ćźšćžŠæ ‡ç­Ÿçš„æ•°æźéƒœæ˜Żć·Čç»æäŸ›ć„œçš„ă€‚èż™ćœšèż™çŻ‡æ–‡ç« äž­ïŒŒä»–ä»Źè€ƒè™‘ćŠ‚äœ•ćŸșäșŽé€æžćąžć€šçš„æ ‡ç­Ÿæ•°æźæ›Žæ–°HashingæšĄćž‹ç ç»™ç”šæˆ·ćšć‡șćż«é€Ÿç›žćș”ïŒŒèą«ç§°äžșSmart Hashing Update.æ‰€è°“äž»ćŠšć­Šäč ïŒŒć°±æ˜Żçł»ç»Ÿè‡ȘćŠšçš„æŒ‘é€‰äž€äș›æ•°æźç»™ç”šæˆ·èż›èĄŒæ ‡èź°ïŒŒç„¶ćŽćŸșäșŽć·Čç»ć­˜ćœšçš„æ•°æźć’Œæ–°æ ‡ć‡†çš„æ•°æźæ›Žæ–°æ•ŽäžȘHashingæšĄćž‹ă€‚ä»–ä»Źçš„çź—æł•æ”çš‹è§äž‹ć›ŸïŒŒæŻæŹĄç”±ç”šæˆ·æ ‡ć‡șæ–°æ•°æźäč‹ćŽïŒŒæ·»ćŠ ćˆ°çŽ°æœ‰æ•°æźé›†äž­ïŒŒç„¶ćŽç”±çł»ç»ŸæŒ‘选那äș›Hashäœéœ€èŠèż›èĄŒæ›Žæ–°ïŒŒèą«æŒ‘ć‡ș的täžȘbit䜍ćŻčćș”çš„Hashć‡œæ•°ć‚äžŽæœŹèœźæ›Žæ–°ïŒŒé‚Łć…¶ćźžćŠ‚äœ•æŒ‘é€‰èż™täžȘbitäœæŻ”èŸƒć…łé”źïŒŒæœŹæ–‡æ˜Żç»™äș†äž€ç§ç­–ç•„1Consistency-based Selection2Similarity-based SelectionćŸșäșŽäž€è‡Žæ€§é€‰æ‹©æ˜Żè€ƒè™‘æ•Žäœ“æ•°æźé›†ć±žäșŽćŒäž€ç±»çš„Hashç æŻäž€äœäžŠçš„äž€è‡Žæ€§æ˜ŻćŠæŻ”èŸƒćŒșïŒŒćˆ€æ–­ćŒæ ·äœçš„æ ‡ç­Ÿ{-1,1}æ˜ŻćŠæŻ”èŸƒäž€è‡ŽïŒŒæ˜ŻćŠéƒœæ˜Żæ­Łäž€ïŒŒæˆ–éƒœæ˜ŻèŽŸäž€ïŒŒćŠ‚æžœäž€è‡Žæ€§äžć„œçš„èŻïŒŒæˆ‘ä»Źć°±æŠŠćźƒæŒ‘é€‰ć‡șæ„ć‚äžŽæ›Žæ–°ïŒ›ćœ“ç„¶èż™ç§ç­–ç•„çš„çŒșç‚čć°±æ˜ŻæČĄæœ‰è€ƒè™‘ć†…éƒšæ•°æźć’Œć€–éƒšæ–°æ•°æźçš„ç›žäŒŒæ€§ïŒŒć› è€ŒçŹŹäșŒç§æ˜ŻćŸșäșŽç›žäŒŒćșŠé€‰æ‹©ïŒŒćșŠé‡ćŒäž€ç±»ćˆ«ć†…çš„HashçŒ–ç æ•ˆæžœć„œäžć„œïŒŒCVPR2012䞊给ć‡șäș†äž€äžȘ性胜ćșŠé‡æŒ‡æ ‡ć…ŹćŒïŒŒHæ˜ŻćŒäž€ç§ç±»ćˆ«çš„Hash码Sæ˜Żć…łè”çŸ©é˜”ïŒŒèż™äžȘæŒ‡æ ‡è¶Šć°çš„èŻïŒŒèŻŽæ˜Žæ•ˆæžœæŻ”èŸƒć„œă€‚èż™èŸčäžșäș†æŒ‘ć‡șæ•ˆæžœäžć„œçš„täžȘHashć‡œæ•°ćŻčèż™äžȘæŒ‡æ ‡èż›èĄŒäș†ć˜ćœąïŒŒäŸæŹĄæŠŠçŹŹk䜍从Hashć‡œæ•°äž­ć‰”é™€ïŒŒćŻčæŻ”ć‰”é™€ć“ȘäžȘHash䜍äč‹ćŽèż™äžȘæŒ‡æ ‡äž‹é™æŻ”èŸƒæ˜Žæ˜Ÿçš„èŻïŒŒć°±æŠŠæłąćŠšćœ±ć“æŻ”èŸƒć€§çš„täžȘ挑选ć‡șæ„èż™ć°±ćźŒæˆäș†æŒ‘é€‰ć·„äœœïŒŒç„¶ćŽæ čæźćžŠæ ‡ç­Ÿæ•°æźèż›èĄŒé‡æ–°ć­Šäč 
  23. ACLäžŠæœ‰äž€çŻ‡æ˜ŻćšMT的CVPRäžŠèż˜æœ‰äž€çŻ‡æ˜ŻćšćŒ»ć­Šć›Ÿç‰‡æŁ€çŽąçš„ă€‚ć‰©äž‹çš„èźș文ć€Șć€šïŒŒæˆ‘äčŸć°±æČĄæœ‰äž€äž€ćˆ—ć‡șæ„ă€‚