Академический Документы
Профессиональный Документы
Культура Документы
2005-12-13
Contents
Biological Immune System Artificial Immune System Basic Algorithms of AIS AIS design procedure Case Studies
Malicious Executable Detection Film Recommender
Robustness Feature extraction Diversity Reinforcement learning Memory; Dynamically changing coverage Distributed Multi-layered Adaptive
Y. Tan---Artificial Immune Sys. 4
2005-12-13
2005-12-13
Immune cells
There are two primarily types of lymphocytes:
B-lymphocytes (B cells) T-lymphocytes (T cells)
2005-12-13
Where is it?
P r im a r y l y m p h o i d o r g a n s S e c o n d a r y lym p h o id o r g a T o n s ils a n d a d e n o id s
T hym us S p le e n
2005-12-13
Lym phocytes
Antigen
Substances capable of starting a specific immune response commonly are referred to as antigens This includes some pathogens such as viruses, bacteria, fungi etc .
2005-12-13
vs
Acquired
Cell Mediated
vs
Humoral
Antibody
Y. Tan---Artificial Immune Sys. 10
2005-12-13
( IV A c tiv a te d T - c e ll
L y m p h o k in e s ( V I )
A c t iv a t e d ( p la s m a
B - c e ll c e ll)
( V II )
2005-12-13
11
Self/Non-Self Recognition
Immune system needs to be able to differentiate between self and non-self cells Antigenic encounters may result in cell death, therefore
Some kind of positive selection Some element of negative selection
2005-12-13
12
2005-12-13
The immune recognition is based on the complementarity between the binding region of the receptor and a portion of the antigen called epitope. Antibodies present a single type of receptor, antigens might present several epitopes. This means that each antibody can recognize a single antigen
Y. Tan---Artificial Immune Sys.
13
Clonal Selection
Clonal deletion (negative selection) Self-antigen Proliferation (Cloning)
M
Memory cells
Plasma cells
Foreign antigens
2005-12-13
14
2005-12-13
16
Continuous learning
2005-12-13 Y. Tan---Artificial Immune Sys. 17
Learning (2)
Primary Response Antibody Concentration Secondary Response Cross-Reactive Response
Lag Lag
Lag
...
Response to Ag2
... ...
Antigen Ag1 Antigens Ag1, Ag2
...
Antigen Ag1 + Ag3 Time
2005-12-13
18
Back
Define host (body cells) from external entities. When an entity is recognized as foreign (or dangerous)- activate several defense mechanisms leading to its destruction (or neutralization). Subsequent exposure to similar entity results in rapid immune response. Overall behavior of the immune system is an emergent property of many local interactions.
2005-12-13
19
Back
Immune metaphors
Other areas Idea! Idea
AI model Formal logic Formal linguistic Neural computing (NC) Neural networks (NN) Cellular automata (CA) Artificial immune systems (AIS) Genetic Algorithms (GA)
22
Some History
Developed from the field of theoretical immunology in the mid 1980s.
Suggested we might look at the IS
1990 Bersini first use of immune algorithms to solve problems Forrest et al Computer Security mid 1990s Hunt et al, mid 1990s Machine learning More
2005-12-13 Y. Tan---Artificial Immune Sys. 23
AIS Scope
Pattern recognition; Fault and anomaly detection; Data analysis; Data mining (classification/clustering) Agent-based systems; Scheduling; Machine-learning; Autonomous navigation and control; Search and optimization methods; Artificial life; Security of information systems; Optimization; Just to name a few.
Y. Tan---Artificial Immune Sys. 24
2005-12-13
Back
Computer Security(Forrest949698, Kephart94, Lamont9801,02, Dasgupta9901, Bentley0001,02) Anomaly Detection (Dasgupta960102) Fault Diagnosis (Ishida9293, Ishiguro94) Data Mining & Retrieval (Hunt9596, Timmis9901, 02) Pattern Recognition (Forrest93, Gibert94, de Castro 02) Adaptive Control (Bersini91) Job shop Scheduling (Hart98, 01, 02) Chemical Pattern Recognition (Dasgupta99) Robotics (Ishiguro9697,Singh01) Optimization (DeCastro99,Endo98, de Castro 02) Web Mining (Nasaroui02,Secker05) Fault Tolerance (Tyrrell, 01, 02, Timmis 02) Autonomous Systems (Varela92,Ishiguro96) Engineering Design Optimization (Hajela96 98, Nunes00)
2005-12-13 Y. Tan---Artificial Immune Sys. 25
2005-12-13
26
A3
B2
C8
D5
A3 B2 C8 D5 Expressed Ab molecule
2005-12-13
27
No
P ro te c te d S trin g s (S )
M a tc h Yes N o n -s e lf D e te c te d
No
Monitoring 28
2.1 Antigenic binding: determine affinity to each P 2.2 Affinity maturation: select n highest affinity from P and clone and mutate prop. to affinity with Ag, then add new mutants to P
3. Metadynamics: 3.1 select highest affinity P to form part of M 3.2 replace n number of random new ones 4. Cycle: repeat 2 and 3 until stopping criteria (e.g. Max Generation)
2005-12-13 Y. Tan---Artificial Immune Sys. 29
Agj Ab{d} Abj* Ab {r} Ab {m} fj Select Select Fj* Ab {n} Cj* Clone
L.N. de Castro, et.al., Learning and optimization using the clonal selection principle, IEEE Trans. Evolutionary computation, vol.6, no.3, June 2002, pp.239251
Select
Y. Tan---Artificial Immune Sys.
Cj
30
2005-12-13
2005-12-13
31
Back
Somatic Hypermutation
Mutation rate in proportion to affinity Very controlled mutation in the natural immune system Trade-off between the normalized antibody affinity D* and its mutation rate ,
1 0 .9 0 .8 0 .7 0 .6
0 .5 0 .4 0 .3 0 .2 0 .1 0 0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 1
1 0 = 2 0
2005-12-13
D *
34
Problem
2005-12-13
Application Domain
Y. Tan---Artificial Immune Sys. 35
A n t ib o d y
2005-12-13
36
Representation
Vectors Ab = Ab1, Ab2, ..., AbL Ag = Ag1, Ag2, ..., AgL Real-valued shape-space Integer shape-space Binary shape-space Symbolic shape-space
Y. Tan---Artificial Immune Sys. 37
2005-12-13
D=
2 ( Ab Ag ) i i i =1
Other distance measures such as Hamming, Manhattan etc. etc. Affinity Threshold
2005-12-13
38
39
Back
AIS Design
Problem description Deciding the immune principles used for problem solving Engineering the AIS
Defining the types of immune components used Defining the representation for the elements of the AIS Applying immune principle to problem solving The meta-dynamics of an AIS
Back
New!
Immuneocomputing -- IC
By Tarakanov, A. 2001. Aims of A proper mathematical framework; A new kind of computing; A new kind of hardware. New concepts of formal protein (FP) ------- vs. neuron formal immune networks (FIN)------- vs. NN
Refer to
2005-12-13
A.O. Tarakanov, V.A. skormin, and S.P. Sokolova, Immunocomputing: Principles and Applications, Springer, 2003.
Y. Tan---Artificial Immune Sys. 42
2005-12-13
43
New!
In the danger model, the idea is to recognise danger rather than non self. The screening is accomplished post production through an external danger signal. Thus the production of autoreactive antibodies (which react to self) is allowed. If an (e.g. autoreactive) antibody matches a stimulus in the absence of danger, it is removed. Thus harmless antigens are tolerated, and changing self accommodated.
Matzinger (2002). The Danger Model: A renewed sense of self , Science 296: 301-304.
2005-12-13 Y. Tan---Artificial Immune Sys. 44
Danger Zone
Stimulation Danger Zone Antibodies Antigens Cells Damaged Cell Danger Signal
2005-12-13 Y. Tan---Artificial Immune Sys. 46
Danger = Context ?
It could also be useful for data mining, where the danger signal is a proxy measure of interest Danger Zone can be spatial or temporal
Andrew Secker, Alex Freitas, and Jon Timmis (2005) Towards a danger theory inspired artificial immune system for web mining in A Scime, editor, Web Mining: applications and techniques, pages 145-168 (Idea Group)
2005-12-13 Y. Tan---Artificial Immune Sys. 48
Back
Anjum Iqbal, Mohd Aizaini Maarof, Danger Theory and Intelligent Data Processing, International Journal of Information Technology, Vol.1, No.1, 2004. Andrew Secker, Alex A. Freitas, and Jon Timmis, A Danger Thory Inspired Approach to Web Mining, Computing Lab. University of Kent, Canterbury, Kent, UK.2005 So on.
2005-12-13 Y. Tan---Artificial Immune Sys. 49
The Future
More formal approach required? Wide possible application domains. What makes the immune system unique? More work with immunologists:
Danger theory. Idiotypic Networks. Self-Assertion.
2005-12-13 Y. Tan---Artificial Immune Sys. 50
New Events:
Special Session on Artificial Immune Systems at the Congress on Evolutionary Computation (CEC), December 8-12, 2003, Canberra, Australia. Special Session on Immunity-Based Systems at Seventh International Conference on Knowledge-Based Intelligent Information & Engineering Systems (KES), September 3-5, 2003, University of Oxford, UK. Second International Conference on Artificial Immune Systems (ICARIS), September 1-3, 2003, Napier University, Edinburgh, UK. Tutorial on Artificial Immune Systems at 1st Multidisciplinary International Conference on Scheduling: Theory and Applications (MISTA), 12 August 2003, The University of Nottingham, UK. Tutorial on Immunological Computation at International Joint Conference on Artificial Intelligence (IJCAI), August 10, 2003, Acapulco, Mexico. Special Track on Artificial Immune Systems at Genetic and Evolutionary Computation Conference (GECCO), Chicago, USA, July 12-16, 2003
2005-12-13 Y. Tan---Artificial Immune Sys. 52
AIS Resources
Artificial Immune Systems and Their Applications by D Dasgupta (Editor), Springer Verlag, 1999. Artificial Immune Systems: A New Computational Intelligence Approach by L de Castro, J Timmis, Springer Verlag, 2002. Immunocomputing: Principles and Applications by A Tarakanov et al, Springer Verlag, 2003. Third International Conference on Artificial Immune Systems (ICARIS), September 13-16, 2004, University of Catania, Italy. 4th International Conference on Artificial Immune Systems(ICARIS), 14th-17th August, 2005 in Banff, Alberta, Canada
2005-12-13
53
First Page
Thats all
2005-12-13
54
Case Study 1:
trier.de)
Outline
Definition of Terms Goal and Motivation Previous Research works Immune Principle for Malicious Executable Detection Malicious Executable Detection Algorithm Experiments and Discussion Concluding Remarks
2005-12-13 Y. Tan---Artificial Immune Sys. 56
Back
Definition of Terms
Malicious Executable is generally defined as a program that has some malicious functions, such as compromising a systems security, damaging a system or obtaining sensitive information without the permission of users. It includes virus, trojan horse, worm etc. Benign Executable is a normal program without any malicious function.
2005-12-13
57
attempt to detect these new malicious programs with heuristics by hand (costly and ineffective)
Dos/Win32 viruses
Trojan horses
Malicious executables
2005-12-13 Y. Tan---Artificial Immune Sys.
Back
2005-12-13
59
Back
Aiming at developing an automatic detection approach of new malicious executables. Aiming at trying to use artificial immune system (AIS) and artificial neural networks (ANN), to detect malicious executable with a high Detection Rate (DR) with low False Positive Rate (FPR) over others.
2005-12-13 Y. Tan---Artificial Immune Sys. 60
Back
2005-12-13
61
Back
Signature-based Methods
It creates a unique tag for each malicious program so that future examples of it can be correctly classified with a small error rate. And relies on signatures of known malicious executable to generate detection models. Drawbacks: Can not detect unknown and mutated viruses. As increase of the number and type of viruses, its detection speed become slow dramatically. At the same time, the analysis of the signatures of viruses become very difficult, in particular, for the encrypted signatures. (refer to IBM Anti-virus Groups report: R.W. Lo, K.N. Levitt, and R.A. Olsson. MCF: a Malicious Code Filter. Computers & Security, 14(6):541566., 1995.)
2005-12-13 Y. Tan---Artificial Immune Sys. 62
Back
Drawbacks: Time-consuming analysis method. Only discover some unknown viruses, but its false detection rate is very high.
For detecting unknown virus based on ANN, IBM Anti-virus Group also proposes one method to detect Boot Sector viruses only.
(refer to W. Arnold and G. Tesauro. Automatically Generated Win32 Heuristic Virus Detection. Proceedings of the 2000 International Virus Bulletin Conference, 2000.)
2005-12-13 Y. Tan---Artificial Immune Sys. 63
Back
M.G. Schultz developed a framework that used data mining algorithms, i.e., Multi-Nave Bayes method, to train multiple classifiers on a set of malicious and benign executables to detect new examples (unknown ME).
(refer to M.G. Schultz.,E. Eskin and E. Zadok . Data Mining Methods for Detection of New Malicious Executables. IEEE Symposium on Security and Privacy, May 2001.)
2005-12-13
64
2005-12-13
65
Computing model Formal logic Formal linguistic Artificial Neural networks (ANN) Cellular automata (CA) Artificial immune systems (AIS) Genetic Algorithms (GA)
66
Molecular Molecular
Knowledge Storage
2005-12-13
67
Back
Back
Y
self non-self
2005-12-13
70
Back
2005-12-13
71
Detectors
2005-12-13
73
back
Detector Representation 1
Detector Representation 2
Detector Representation 3
Combination of detectors
2005-12-13 Y. Tan---Artificial Immune Sys. 74
Back
Extracting property
anomaly
Classifier
Output
2005-12-13
76
Back
Extraction of Anomaly Characteristics -Non-self Thickness (NST) Non-self Detection NST, as Anomaly Property, is defined as the ratio of number of non-self units to file binary sequence, pl=nn/(nn+ns). If there are m kinds of detectors, the file has a NST Vector P=(pl1, pl2, , plm)T.
2005-12-13
79
Is Nonself ? N ns add 1
nn add 1
End
2005-12-13 Y. Tan---Artificial Immune Sys. 80
Back
BP Network Classifier
We use Anomaly Property Vector (APV), i.e., NST vector P, as input variable of two-layer BP network classifier. The number of nodes of input layer equals to APVs dimension. The Sigmoid transfer function is chosen for the hidden layer and Linear function for the output layer.
2005-12-13 Y. Tan---Artificial Immune Sys. 82
Back
pl1 pl2
P
plm
2005-12-13
83
Back
2005-12-13
84
Back
Back
Eg(b) is Gene of generating detector, ld {16 24326496}, and lstep=8bits. By using the detector generating algorithm, we can get D16, D24, D32, D64, and D96, separately.
Table1: Detectors generation result
Code Length ld |Dld| store structure
2005-12-13
16 65536
24
32
64
96
86
NST p24 File No. (a) NST of files, where symbol x represents benign program (Red), malicious program (Blue)
2005-12-13
87
P32
NST p32
(a) NST of files, where symbol x represents benign program, malicious program
2005-12-13 Y. Tan---Artificial Immune Sys.
88
P64
2005-12-13
NST p64
(a) NST of files, where symbol x represents benign program (Red), malicious program (Blue)
89
Back
C o d e le n g th l d b its
2005-12-13
Note: from the bottom to up, the FPR Y. Tan---Artificial Immune Sys. is 0%, 0.5%, 1%, 2%, 4%, 8%, and 16%, in sequence.
91
24
bit
32bits
(a) NST of files for mixture of D24, D32 and D64. x benign program (in Red), malicious program (in Blue).
2005-12-13
(b) ROC Curve of mixed detector set of D24, D32, D64 and D96
93
60 40 20 0 0 2
10
12
Back
Algorithm Complexities
Operation type 1 Algorithm Name detectors Prob. Info. Amount ltrain >>ltrain Operation type 2 Name detector matching Searching P(Fi/C) Amount 80ltes
t
Store Space
MEDA Bayes
0.4Gb 1Gb
2005-12-13
95
Remarks
Back
For short binary sequence and single detector set for the detection of malicious executables, the performance of D24 is the best, giving out DR 80.6% with FPR 3%. For long code length of detector and multidetector set, our method obtains the best performance of DR 97.46% with FPR 2%, over current methods. This result verifies
diversity of detector representation can decrease anomaly detection holes. non-self thickness detection.
2005-12-13 Y. Tan---Artificial Immune Sys.
Back
96
Case Study 2:
Film Recommender
From Dr. Dr Uwe Aickelin (http://www.aickelin.com) University of Nottingham, U.K.,
z
Prediction:
What rating would I give a specific film?
Recommendation:
Give me a top 10 list of films I might like.
2005-12-13
97
2005-12-13
at io
User Profile: set of tuples {movie, rating} Me: My user profile. Neighbour: User profile of others. Affinity metric: Correlation score.
Su pp re ss io
ul im
St
Group of antibodies similar to antigen and dissimilar to other antibodies Weighted Score based on Similarities.
2005-12-13 Y. Tan---Artificial Immune Sys.
2005-12-13
100
We do not have users votes for every film. We want to predict the vote of user u4 on movie m3.
2005-12-13 Y. Tan---Artificial Immune Sys. 101
AIS
u4
Ag
102
u2,u3 Ab1
Ag Ab2 Ab3
103
2005-12-13
2005-12-13
105
2005-12-13
106
Recommendation:
4 out of 5 films correct (AIS). 3 out of 5 films correct (Pearson). Back
2005-12-13 Y. Tan---Artificial Immune Sys. 107