Artificial Immune Systems: A New Computational Intelligence Approach

New Trends in Intelligent Information Processing and Web Mining. Zakopane, Poland, June 2-5, 2003
Jonathan Timmis Computing Laboratory University of Kent CT2 7NF. UK. J.Timmis@kent.ac.uk http:/www.cs.kent.ac.uk/~jt6

Novel paradigms are proposed and accepted not necessarily for being faithful to their sources of inspiration, but for being useful and feasible

What do I want to achieve?

Give you a taster of what AIS is all about
Define an AIS Why do we find the immune system useful? Explain what AIS are Show you where they are being used Some high level case studies Comments for the future

I wont
Talk about all areas of AIS and applications Talk too much about how AIS relate to other bioinspired ideas (although I will mention it) Go into too much detail: this is an introduction

Outline
What are AIS? Useful immunology Thinking about AIS Application Areas and Case Studies The Future

Why the Immune System?

Recognition
Anomaly detection Noise tolerance

Robustness Feature extraction Diversity Reinforcement learning Memory; Dynamically changing coverage Distributed Multi-layered Adaptive

A Definition
AIS are adaptive systems inspired by theoretical immunology and observed immune functions, principles and models, which are applied to complex problem domains

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 algos to solve problems Forrest et al Computer Security mid 1990s Hunt et al, mid 1990s Machine learning

Scope of AIS:
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)

Scope of AIS (Cont):

Job shop Scheduling (Hart98, 01, 02) Chemical Pattern Recognition (Dasgupta99) Robotics (Ishiguro9697,Singh01) Optimization (DeCastro99,Endo98, de Castro 02) Web Mining (Nasaroui02) Fault Tolerance (Tyrrell, 01, 02, Timmis 02) Autonomous Systems (Varela92,Ishiguro96) Engineering Design Optimization (Hajela96 98, Nunes00) And so on

Outline
What are AIS? Useful immunology Thinking about AIS Application Areas and Case Studies The Future

Role of the Immune System

Protect our bodies from pathogen and viruses Primary immune response
Launch a response to invading pathogens

Secondary immune response

Remember past encounters Faster response the second time around

How does it work: A simplistic view

M H C A P C p r o te in A n tig e n ( I ) P e p tid e ( I I ) T - c e ll ( I I I ) B - c e ll ( V )

( I V A c tiv a te d T - c e ll

L y m p h o k in e s ( V I )

A c tiv a te d ( p la s m a

B - c e ll c e ll)

( V I I )

Immune cells
There are two primarily types of lymphocytes:
B-lymphocytes (B cells) T-lymphocytes (T cells)

Others types include macrophages, phagocytic cells, cytokines, etc.

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

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 .

Immune Pattern Recognition

BCR or Antibody

B-cell Receptors (Ab) Epitopes Antigen

B-cell

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

Clonal Selection
Clonal deletion (negati e selection) S lf- nti n

Antibody Selection Differentiation Pl s

For i

ti

S lf- nti

Clonal deletion (negati e selection)

Proliferation (Cloning)

o r c lls

c lls

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

Immune Network Theory

Idiotypic network (Jerne, 1974) B cells co-stimulate each other Treat each other a bit like antigens Creates an immunological memory
P ra p a to e S p re s n u p s io N g tiv re p n e ea e sos Ag 1 2 Id to e io p A tib d n oy A tiv tio c a n P s ere p n e o itiv sos 3

Reinforcement Learning and Immune Memory

Repeated exposure to an antigen throughout a lifetime Primary, secondary immune responses Remembers encounters
No need to start from scratch Memory cells

Continuous learning

Learning (2)
Pri ary Res onse Secondary Res onse Cross-Reacti e Res onse

Antibody Concentration

Lag Lag

Lag

Res onse to Ag1 Res onse to Ag1

...
Res onse to Ag2

Res onse to Ag1 + Ag3

... ...
Antigen Ag1 Antigens Ag1, Ag2

...
Antigen Ag1 + Ag3 Ti e

Immune System: Summary

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. So it is useful?

Outline
What are AIS? Useful immunology Thinking about AIS Application Areas and Case Studies The Future

Artificial Immune Systems

AIS are adaptive systems inspired by theoretical immunology and observed immune functions, principles and models, which are applied to complex problem domains

This Section
General Framework for describing and constructing AIS models A short review of where AIS are used today
Can not cover them all, far too many Also we are not experts in all application areas !

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 components Be able to describe their interactions Quite high level abstractions Capture general purpose processes that can be applied to various areas

General Framework for AIS

Immune Algorithms Affinity Measures Representation Application Domain

Representation Shape Space

Describe the general shape of a molecule
Antigen

Antibody

Describe interactions between molecules Degree of binding between molecules

Representation
Vectors Ab = Ab1, Ab2, ..., AbL Ag = Ag1, Ag2, ..., AgL Real-valued shape-space Integer shape-space Binary shape-space Symbolic shape-space

Define their Interaction

Define the term Affinity Distance measures such as Hamming, Manhattan etc. etc. Affinity Threshold

Basic Immune Models and Algorithms

Negative Selection Algorithms Clonal Selection Algorithm Immune Network Models Somatic Hypermutation

Negative Selection (NS) Algorithms

Forrest 1994: Idea taken from the negative selection of T-cells in the thymus Applied initially to computer security Split into two parts:
Censoring Monitoring
Self strings (S)
D e te c to r S e t (R )

Generate random strings (R0)

Match No Yes Reject

Detector Set (R)

Pro te c te d S trin g s (S )

M a tc h Y es N o n -s e D e te c te d

No

 

Clonal Selection Algorithm (de Castro & von Zuben, 2001)

1. Initialisation: Randomly initialise a population (P) 2. Antigenic Presentation: for each pattern in Ag, do: 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

Discrete Immune Network Models (Timmis & Neal, 2001)

1. 2. Initialisation: create an initial network from a sub-section of the antigens Antigenic presentation: for each antigenic pattern, do: 2.1 Clonal selection and network interactions: for each network cell, determine its stimulation level (based on antigenic and network interaction) 2.2 Metadynamics: eliminate network cells with a low stimulation 2.3 Clonal Expansion: select the most stimulated network cells and reproduce them proportionally to their stimulation 2.4 Somatic hypermutation: mutate each clone 2.5 Network construction: select mutated clones and integrate 3. Cycle: Repeat step 2 until termination condition is met

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 E,
1 0 .9 0 .8 0 .7 0 .6

V = 5

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

V = 10 V = 20

D*

Case Study: Data Mining

Data mining: Problem description

More benchmark problem in this case Assume a set of labelled vectors Classification

AIRS: (Artificial Immune Recognition System) Watkins 2003

Clonal Selection Based initially on immune networks, though found this did not work Resource allocation Somatic hypermutation
Eventually

Antibody/antigen binding

AIRS: Mapping from IS to AIS

Antibody Recognition Ball Antigens Immune Memory Feature Vector Combination of feature vector and vector class Training Data Memory cellsset of mutated ARBs

Classification
Stimulation of an ARB is based not only on its affinity to an antigen but also on its class when compared to the class of an antigen Allocation of resources to the ARBs also takes into account the ARBs classifications when compared to the class of the antigen Memory cell hyper-mutation and replacement is based primarily on classification and secondarily on affinity

AIRS Algorithm
Data normalization and initialization Memory cell identification and ARB generation Competition for resources in the development of a candidate memory cell Potential introduction of the candidate memory cell into the set of established memory cells

AIRS: Performance Evaluation

Fishers Iris Data Set Pima Indians Diabetes Data Set

Ionosphere Data Set

Sonar Data Set

Classification Accuracy
Important to maintain accuracy
AIRS1: Accuracy AIRS2: Accuracy Iris 96.7 96.0

Ionosphere

94.9

95.6

Diabetes

74.1

74.2

Sonar

84.0

84.9

Features
No need to know best architecture to get good results Default settings within a few percent of the best it can get User-adjustable parameters optimize performance for a given problem set Generalization and data reduction

aiNET: Artificial Immune Network for Data Mining

Problem description
More benchmark problem in this case Assume a set of unlabelled vectors We can ask the questions:
Is there a large amount of redundancy? Are there any groups or subgroups intrinsic to the data? What is the structural or spatial distribution?

aiNET: Immune principles employed

B-cells (antibodies) Antigens Antibody/antigen binding Clonal selection process Immune network theory Combined with statistical analysis tools

Data mining: Immune Network Algorithm

1. Initialization: create an initial random population of network antibodies; 2. Antigenic presentation: for each antigenic pattern, do:
2.1 Clonal selection and expansion: 2.2 Affinity maturation: 2.3 Clonal interactions: 2.4 Clonal suppression: 2.5 Metadynamics: 2.6 Network construction: 3. Network interactions:

4. Network suppression: 5. Diversity: 6. Cycle: repeat Steps 2 to 4 until a pre-specified number of iterations is reached.

Data mining: Mapping from IS to aiNET

Immune System B-cell (antibody) Antigen Binding Cell cloning Somatic hypermutation Immune network Metadynamics aiNET Internal data vector Training data vector Calculation of Euclidean distance Duplication of internal data vectors Affinity proportional mutation Network of internal data vectors Removal and creation of internal data vectors

Data mining: Clustering (aiNet)

Limited visualisation Interpret via MST or dendrogram Compression rate of 81% Successfully identifies the clusters
Training Pattern
T ra in in g P a tte rn s
1 1 11 1 1 11 1 11 1 1 1 11 11 1 11 11 1 1 1 111 1 1 1 1 1 11 11 11 1 111 1 11 1 1 111 11 11 1 11 11 11 111 1 11 11 11 11 11 1 1 11 1 1 1 1 1 5 55 1 1 1 1 2 22 11 11 1 2 1 1 1 1 1 1 1 1 22 22 22 1 1 5 5 55 1 555 5 2 22 11 11 1 22222 2 2 5 555 1 1 5 55 2 22222 2 22 555 5 5 222 2 1 1 1 1 22 11 6 66 5555555 5 11 6 6 1 11 2222 6 5555 5 5 5 5 2 2 2 2 1 6 55555 5 555 5 2 1 1 22 22 1 1 1 55 5 5 5 5 5 22222 1 16 1 22222 2 22 2 22 55 5 5 5 5 55 1 1 1 1 1 22 1 22 2 2 1 11 1 1 1 5 5 55 1 5 1 5555 5 5 55 1 2 22 1 1 11 1 1 11 11 11 11 1 1 1 11 11 1 111 11 1 1 11 11 1 1 1 1 11 11 1 11 1 1 1 1 11 1 11 11 1 11 1 11 111 1 1 1 1 111 1 11 1 1 111 1 1 1 1 1 1 1 11 4 4 4 4 4 44 44 4 4 444 4 4 44 44 44 4 4 4444444 44 4 44444 4 44 4 4 444 4 44 4 4 4 44 4 4 44 4 444 4444 4 4 44 4 44 4 4 44 444 4 44 3 3 33 3 33 3 3 3 33 3 3 3 3 33 3 3 3 3 33 33 3333 3 3 3 33 3 33 3 33 3 33 33 3 33 3 3 3 33 333333 3 3 333 3 3 3 3333 3 3 3 3 3 3 3 33 3 3 3 33 33 3 0 .8 8 8 88 8 88 8 88 8 8 8 88 8 8 8 8 88

0 .6

0 .4

0 .2

7 7 777 8 8 7 7 7 8 8 8 7 88 8 7 7 8 7 77 7 7 888 88 8 7 7 7 8 7 7 8 8 7 88 8 7 8 77 7 88 8 8 77 77 88 88 8 8 77 8 8 77 77 7 7 8 7 7 0 .4 0 .6 0 .8

7 77 7 7 77 7

0 .2

Result immune network

Data mining:Hierarchical Clustering (aiNET)

1-1

1-2

1-3

1-4

1-5

2-1

2-2

2-3

2-4

Other Interesting Applications

Immune Network for continuous learning (Neal 2002)
Track moving data over time Maintains clusters in absence of patterns Useful for dynamic environments

Continuous Classification
Email classification of interesting/non-interesting emails Changing profile of the user Maintain classification accuracy Comparable to Nave Bayes

New Trends
Danger Theory
Not self/non-self but Danger/Non-Danger Immune response is initiated in the tissues. Danger Zone. This makes it context dependant

Could this be useful for Web Mining?

Summary
Covered much, but there is much work not covered (so apologies to anyone for missing theirs)

Immune metaphors
Antibodies and their interactions Immune learning and memory Self/non-self
Negative selection

Application of immune metaphors

The Future
Rapidly emerging field Much work is very diverse
Framework helps a little More formal approach required?

Wide possible application domains What is it that makes the immune system unique? More work with immunologists
Theories such as Danger theory, Self-Assertion may have something to say to AIS

The Future (2)

ARTIST: A Network for Artificial Immune Systems (EPSRC funded network) Work towards:
A theoretical foundation for AIS as a new CI Extraction of accurate metaphors Immune System Modelling Application of AIS

Train PhD students Fund workshops/meetings Coordinate and Disseminate UK based AIS research (links to Europe)

The Future (hopefully)

IT IS: Information Technology Inspired by the Immune System FP 6 IP:
16 institutions across Europe Create a European Library of immune algorithms Theoretical analysis of AIS Application of AIS
Autonomous boat Immunoinformatics Web Mining

Modelling of Immune System

AIS Resources: Books

Artificial Immune Systems and Their Applications by Dipankar Dasgupta (Editor) Springer erlag, January 1999. Artificial Immune Systems: A New Computational Intelligence Approach by Leandro N. de Castro, Jonathan Timmis, Springer erlag, November 2002. Immunocomputing: Principles and Applications by Alexander O. Tarakanov, ictor A. Skormin, Svetlana P. Sokolova, Springer erlag, April 2003.

AIS Related Events in 2003:

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