View
217
Download
2
Category
Preview:
Citation preview
An Introduction to Artificial Immune Systems
Dr. Jonathan Timmis
Computing Laboratory
University of Kent at Canterbury
CT2 7NF. UK.
J.Timmis@ukc.ac.uk
http:/www.cs.ukc.ac.uk/people/staff/jt6
ES2001Cambridge. December 2001.
Overview of TutorialWhat are we going to do?:First Half:
Describe what is an AISWhy bother with the immune system?Be familiar with relevant immunology
Second Half:Appreciation of were AIS are usedBe familiar with the building blocks of AIS
Resources
Immune metaphors
Immune System
Idea! Idea ‘
Other areas
Artificial Immune Systems
Why the Immune System?Recognition
Anomaly detectionNoise tolerance
RobustnessFeature extractionDiversityReinforcement learningMemoryDistributedMulti-layeredAdaptive
Artificial Immune Systems
AIS are computational systems inspired by theoretical immunology and observed immune functions, principles and models, which are applied to complex problem domains (de Castro & Timmis, 2001)
Some History
Developed from the field of theoretical immunology in the mid 1980’s.
Suggested we ‘might look’ at the IS
1990 – Bersini first use of immune algos to solve problemsForrest et al – Computer Security mid 1990’sHunt et al, mid 1990’s – Machine learning
Scope of AISFault and anomaly detectionData Mining (machine learning, Pattern recognition)Agent based systemsSchedulingAutonomous controlOptimisationRoboticsSecurity of information systems
Part I – Basic Immunology
Role of the Immune System
Protect our bodies from infection
Primary immune responseLaunch a response to invading pathogens
Secondary immune responseRemember past encounters
Faster response the second time around
How does it work?
Where is it?
Multiple layers of the immune system
Phagocyte
Adaptive immune
response
Lymphocytes
Innate immune
response
Biochemical barriers
Skin
Pathogens
Immune Pattern Recognition
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 different antibodies can recognize a single antigen
Antibodies
Antigen binding sites
VH
VL
CH CH
VL
CL
CH
VH
CH
CL
Fc
FabFab
Antibody Molecule Antibody Production
Clonal Selection
Main Properties of Clonal Selection (Burnet, 1978)
Elimination of self antigens
Proliferation and differentiation on contact of mature lymphocytes with antigen
Restriction of one pattern to one differentiated cell and retention of that pattern by clonal descendants;
Generation of new random genetic changes, subsequently expressed as diverse antibody patterns by a form of accelerated somatic mutation
T-cells
Regulation of other cells
Active in the immune responseHelper T-cells
Killer T-cells
Reinforcement Learning and Immune Memory
Repeated exposure to an antigen throughout a lifetime
Primary, secondary immune responses
Remembers encountersNo need to start from scratch
Memory cells
Associative memory
Learning (2)
Antigen Ag1 Antigens Ag1, Ag2
Primary Response Secondary Response
Lag
Response to Ag1
Ant
ibod
y C
once
ntra
tion
Time
Lag
Response to Ag2
Response to Ag1
...
...
Cross-Reactive Response
...
...
Antigen Ag1 + Ag3
Response to Ag1 + Ag3
Lag
Immune Network Theory
Idiotypic network (Jerne, 1974)
B cells co-stimulate each otherTreat each other a bit like antigens
Creates an immunological memory
Immune Network Theory(2)
Shape Space Formalism
Repertoire of the immune system is complete (Perelson, 1989)
Extensive regions of complementarity
Some threshold of recognition
V
V
V
V
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
Summary so far ….
Immune system has some remarkable properties
Pattern recognition
Learning
Memory
So, is it useful?
Some questions for you !
Part II –Artificial Immune Systems
This Section
General Framework for describing and constructing AIS
A short review of where AIS are used todayCan not cover them all, far too many
I am not an expert in all areas (earn more money if I was)
Where are AIS headed?
What do want from a Framework?
In a computational world we work with representations and processes. Therefore, we need:
To be able to describe immune system componentsBe able to describe their interactionsQuite high level abstractionsCapture general purpose processes that can be applied to various areas
AIS Framework
De Castro & Timmis, 2002
Immune Representations
Immune Algorithms
Guidelines for developing AIS
Representation – Shape Space
Describe the general shape of a molecule
•Describe interactions between molecules
•Degree of binding between molecules
•Complement threshold
Representation
Vectors
Ab = Ab1, Ab2, ..., AbL
Ag = Ag1, Ag2, ..., AgL
Real-valued shape-space
Integer shape-space
Hamming shape-space
Symbolic shape-space
Define their InteractionDefine the term AffinityAffinity is related to distance
Euclidian
L
iii AgAbD
1
2)(
• Other distance measures such as Hamming, Manhattan etc. etc.
• Affinity Threshold
Basic Immune Models and Algorithms
Bone Marrow Models
Negative Selection Algorithms
Clonal Selection Algorithm
Somatic Hypermutation
Immune Network Models
Bone Marrow ModelsGene libraries are used to create antibodies from the bone marrowAntibody production through a random concatenation from gene librariesSimple or complex libraries
Negative Selection AlgorithmsForrest 1994: Idea taken from the negative selection of T-cells in the thymusApplied initially to computer securitySplit into two parts:
CensoringMonitoring
Negative Selection AlgorithmEach copy of the algorithm is unique, so that each protected location is provided with a unique set of detectorsDetection is probabilistic, as a consequence of using different sets of detectors to protect each entityA robust system should detect any foreign activity rather than looking for specific known patterns of intrusion. No prior knowledge of anomaly (non-self) is requiredThe size of the detector set does not necessarily increase with the number of strings being protectedThe detection probability increases exponentially with the number of independent detection algorithmsThere is an exponential cost to generate detectors with relation to the number of strings being protected (self).
Solution to the above in D’haeseleer et al. (1996)
Clonal Selection Algorithmde Castro & von Zuben, 2001
Randomly initialise a population (P)For each pattern in Ag
Determine affinity to each P’Select n highest affinity from P
Clone and mutate prop. to affinity with Ag
Add new mutants to P endForSelect highest affinity P to form part of MReplace n number of random new ones
Until stopping criteria
Immune Network Models
Timmis & Neal, 2000
Used immune network theory as a basis, proposed the AINE algorithmInitialize AINFor each antigen
Present antigen to each ARB in the AINCalculate ARB stimulation levelAllocate B cells to ARBs, based on stimulation levelRemove weakest ARBs (ones that do not hold any B cells)
If termination condition metexit
elseClone and mutate remaining ARBsIntegrate new ARBs into AIN
Immune Network ModelsDe Castro & Von Zuben (2000c)
aiNET, based in similar principlesAt each iteration step do
For each antigen doDetermine affinity to all network cellsSelect n highest affinity network cellsClone these n selected cells
Increase the affinity of the cells to antigen by reducing the distance between them (greedy search)
Calculate improved affinity of these n cellsRe-select a number of improved cells and place into matrix MRemove cells from M whose affinity is below a set thresholdCalculate cell-cell affinity within the networkRemove cells from network whose affinity is below
a certain thresholdConcatenate original network and M to form new network
Determine whole network inter-cell affinities and remove all those below the set threshold
Replace r% of worst individuals by novel randomly generated onesTest stopping criterion
Somatic HypermutationMutation rate in proportion to affinityVery controlled mutation in the natural immune systemTrade-off between the normalized antibody affinity D* and its mutation rate ,
Part III - Applications
Anomaly DetectionThe normal behavior of a system is often characterized by a series of observations over time. The problem of detecting novelties, or anomalies, can be viewed as finding deviations of a characteristic property in the system.For computer scientists, the identification of computational viruses and network intrusions is considered one of the most important anomaly detection tasks
Virus DetectionProtect the computer from unwanted virusesInitial work by Kephart 1994More of a computer immune system
Detect Anomaly
Scan for known viruses
Capture samples using decoys
Extract Signature(s)
Add signature(s) to databases
Add removal infoto database
Segregatecode/data
AlgorithmicVirus Analysis
Send signals toneighbor machines
Remove Virus
Virus Detection (2)Okamoto & Ishida (1999a,b) proposed a distributed approach Detected viruses by matching self-information
first few bytes of the head of a file the file size and path, etc. against the current host files.
Viruses were neutralized by overwriting the self-information on the infected filesRecovering was attained by copying the same file from other uninfected hosts through the computer network
Virus Detection (3)Other key works include:
A distributed self adaptive architecture for a computer virus immune system (Lamont, 200)Use a set of co-operating agents to detect non-self patterns
Immune System Computational System
Pathogens (antigens) Computer viruses
B-, T-cells and antibodies Detectors
Proteins Strings
Antibody/antigen binding Pattern matching
Security
Somayaji et al. (1997) outlined mappings between IS and computer systemsA security systems need
ConfidentialityIntegrityAvailabilityAccountability Correctness
IS to Security SystemsImmune System Network Environment
Static Data
Self Uncorrupted data
Non-self Any change to self
Active Processes on Single Host
Cell Active process in a computer
Multicellular organism Computer running multiple processes
Population of organisms Set of networked computers
Skin and innate immunity Security mechanisms, like passwords, groups, file permissions, etc.
Adaptive immunity Lymphocyte process able to query other processes to seek for abnormal behaviors
Autoimmune response False alarm
Self Normal behavior
Non-self Abnormal behavior
Network of Mutually Trusting Computers
Organ in an animal Each computer in a network environment
Network Security
Hofmeyr & Forrest (1999, 2000): developing an artificial immune system that is distributed, robust, dynamic, diverse and adaptive, with applications to computer network security.
Kim & Bentley (2001). Hybrid approach of clonal selection and negative selection.
Forrests Model
AIS for computer network security. (a) Architecture. (b) Life cycle of a detector.
Randomly created
Immature
Mature & Naive
Death
Activated
Memory
No match duringtolerization
010011100010.....001101
Exceed
activationthreshold
Don’t exceed
activation threshold
No co stimulation
Co stimulation
Match
Match during
tolerization
Datapath triple
(20.20.15.7, 31.14.22.87, ftp)
Broadcast LAN
ip: 31.14.22.87port: 2000
Internal host
External host
ip: 20.20.15.7 port: 22
Host
sensitivitylevel
secondaryrepresentation
Detectorset
{immature, naive, memory}
Last
activated
matches
0100111010101000110......101010010
Detector
stateActivation
flag
Novelty DetectionImage Segmentation : McCoy & Devarajan (1997)
Detecting road contours in aerial imagesUsed a negative selection algorithm
Hardware Fault Tolerance
Immunotronics (Bradley & Tyrell, 2000)
Use negative selection algorithm for fault tolerance in hardware
Table 4.1.
Immune System Hardware Fault Tolerance
Recognition of self Recognition of valid state/state transition
Recognition of non-self Recognition of invalid state/state transition
Learning Learning correct states and transitions
Humoral immunity Error detection and recovery
Clonal deletion Isolation of self-recognizing tolerance conditions
Inactivation of antigen Return to normal operation
Life of an organism Operation lifetime of a hardware
Machine Learning
Early work on DNA Recognition Cooke and Hunt, 1995
Use immune network theory
Evolve a structure to use for prediction of DNA sequences
90% classification rate
Quite good at the time, but needed more corroboration of results
Unsupervised Learning
Timmis, 2000Based on Hunts work
Complete redesign of algorithm: AINE
Immune metadynamics
Shape space
Few initial parameters
Stabilises to find a core pattern within a network of B cells
Results (Timmis, 2000)
Immune System : AIS
B-cell
B-cell recognition
Immune Network
Somatic Hypermutation
Antigens
Antigen binding
Initial DataArtificial Recognition BallARB NetworkMutation of ARB’s
Training dataMatching between antigen and ARB’s
Another approach
de Castro and von Zuben, 2000aiNET cf. SOFMUse similar ideas to Timmis
• Immune network theory• Shape space
Suppression mechanism different• Eliminate self similar cells under a set threshold
Clone based on antigen match, network not taken into account
Results (de Castro & von Zuben, 2001)
Test Problem Result from aiNET
Supervised Approach
Carter, 2000Pattern recognition and classification system: Immunos-81
Use T-cells, B-cells, antibodies and amino-acid library
Builds a library of data types and classes
Watkins, 2001Resource allocated mechanism (based on network models)
Good classification rates on sample data sets
RoboticsBehaviour Arbitration
Ishiguro et al. (1996, 1997) : Immune network theory to evolve a behaviour among a set of agents
Collective BehaviourEmerging collective behaviour through communicating robots (Jun et al, 1999)Immune network theory to suppress or encourage robots behaviour
Desirable Interacting antibodiescondition and degree of interaction
Action
Paratope Idiotope
SchedulingHart et al. (1998) and Hart & Ross (1999a)Proposed an AIS to produce robust schedules
for a dynamic job-shop scheduling problem in which jobs arrive continually, and the environment is subject to changes.
Investigated is an AIS could be evolved using a GA approach
then be used to produce sets of schedules which together cover a range of contingencies, predictable and unpredictable.
Model included evolution through gene libraries, affinity maturation of the immune response and the clonal selection principle.
DiagnosisIshida (1993) Immune network model applied to the process diagnosis problemLater was elaborated as a sensor network that could diagnose sensor faults by evaluating reliability of data from sensors, and process faults by evaluating reliability of constraints among data.Main immune features employed:
Recognition is performed by distributed agents which dynamically interact with each other;Each agent reacts based solely on its own knowledge; andMemory is realized as stable equilibrium points of the dynamical network.
Comparing Approaches AIS ANN EA
Components Attribute string in S Artificial neurons Strings representing chromosomes
Location of components Dynamic locations Pre-defined/dynamic (deterministic) locations
Dynamic locations
Structure Set of discrete or networked elements
Networked neurons Discrete elements
Knowledge storage Attribute strings/ network connections
Connection strengths Chromosomal strings
Dynamics Learning/evolution Learning Evolution
Metadynamics Elimination/recruitment of components
Constructive/pruning algorithms
Elimination/ recruitment of individuals
Interaction with other components
Through recognition of attribute strings or network connections
Through network connections Through recombination operators and/or fitness function
Interaction with the environment
Recognition of an input pattern or evaluation of an objective function
Input units receive the environmental stimuli
Evaluation of an objective function
Threshold Influences the affinity of elements
Influences neuron activation Influences genetic variations
Robustness Population/network of individuals
Network of individuals Population of individuals
State Concentration and affinity Activation level of output neurons
Genetic information in chromosomes
Control Immune principle, theory or process
Learning algorithm Evolutionary algorithm
Generalization capability
Cross-reaction Network extrapolation Detection of common schemas
Non-linearity Binding activation function Neuronal activation function Not explicit
Characterization Evolutionary and/or connectionist
According to the learning algorithm
Evolutionary
SummaryCovered much, but there is much work not covered (so apologies to anyone for missing theirs)ImmunologyImmune metaphors
Antibodies and their interactionsImmune learning and memorySelf/non-self
• Negative selection
Application of immune metaphors
The Future
Rapidly growing field that I think is very excitingMuch work is very diverse
Framework helps a littleMore formal approach required?
Wide possible application domainsWhat is it that makes the immune system unique?
More Information
http://www.cs.ukc.ac.uk/people/staff/jt6
http://www.msci.memphis.edu/~dasgupta/
http://www.dcs.kcl.ac.uk/staff/jungwon/
http://www.dca.fee.unicamp.br/~lnunes/
http://www.cs.unm.edu/~forrest/
Recommended