Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial...
Transcript of Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial...
![Page 1: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/1.jpg)
Architecture of an Artificial Immune System
Steven A. Hofmeyr, S. Forrest
Presentation by
Ranjani Srinivasan ([email protected])
![Page 2: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/2.jpg)
The Fellowship of the Immune System- The Defenders of the Realm
![Page 3: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/3.jpg)
The Fellowship of the Immune System- The Defenders of the Realm
Courtesy: Kurzgesagt – In a Nutshell, YouTube Channel. The Immune System explained
![Page 4: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/4.jpg)
Properties of the Immune System (IS) Diverse, Distributed, Error Tolerant, Dynamic and Self-monitoring Robustness Adaptable – Recognize and respond to new infections, retain memory Autonomous – No outside control, difficult to impose outside control or inside centralized control. (Exceptions exist)
![Page 5: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/5.jpg)
Basic Components of Immune Response
Localized Interactions – chemical bonding Dynamic system of circulation Decentralized; no hierarchical organization Self-from-Nonself Improve to Harmful Nonself-from-
EverythingElse Two sub-problems
o Detection o Elimination
![Page 6: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/6.jpg)
ARTificial Immune System
ARTIS – Incorporates (most) properties of IS Independent of problem domain Situating in a domain can reduce unnecessary
features or tailor features to problem
![Page 7: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/7.jpg)
Problem definition
Protein Chains – binary strings length l. Disjoint (assumption) subsets of Universe U, S
and N Discrimination or Classification Task Errors – False Positives and False Negatives
![Page 8: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/8.jpg)
Detectors
Modeled after one class – Lymphocytes Combines properties of B-cells, T-cells, and
antibodies Distributed environment modeled as graph
G = (V,E) Affinity to epitopes (region on pathogen) –
approximate string matching r-contiguous bits (More biologically consistent)
![Page 9: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/9.jpg)
Detectors
Activation of lymphocyte – when binding receptors exceeds threshold Modeling Activation Threshold – Match atleast τ strings in given time. Decay match count (ϒ). Once activated, reset count to zero.
![Page 10: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/10.jpg)
Training the Detection System
Negative Selection Algorithm Tolerization – in Thymus. Training set of self Assumption: Self occurs frequently compared
to non-self ARTIS – Distributed Tolerization
![Page 11: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/11.jpg)
Negative Selection Algorithm
![Page 12: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/12.jpg)
Memory
Rapid and efficient secondary response Associative Activated lymphocytes clone; Retain memory
cells Multiple detectors at node in competition.
Closest match – winner. Spread to neighboring nodes Memory detectors have lower activation
thresholds => rapid response
![Page 13: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/13.jpg)
Sensitivity
Cytokines (chemicals) – signal to nearby IS cells Detection node , local sensitivity , Threshold of detectors at is (τ - ) Matches at i go up, sensitivity is increased by 1 Temporal horizon with decay rate ϒw
![Page 14: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/14.jpg)
Co-stimulation
T-cells require second signal of “damage” Model crude approximation of co-
stimulation: human operator Co-stimulation delay T
![Page 15: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/15.jpg)
Lifecycle of Detector
Lympocytes short-lived. Dynamic population Model: pdeath for mature detectors Exception: Memory Detectors. Die only if no
co-stimulation Problem? Limit fraction of memory detectors md
LRU (least Recently used) => Least useful (Is this a valid assumption?)
![Page 16: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/16.jpg)
Lifecycle of Detector
![Page 17: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/17.jpg)
Representation
Population level diversity – MHC. Holes – occur if every match has a self counterpart; No valid detectors can be generated
![Page 18: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/18.jpg)
Representation
Each node with different representation. Modify all incoming detectors
![Page 19: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/19.jpg)
Response
Effector selection – many kinds B-cells – antibodies; Variable and constant
regions Isotype Switching Implementation: Augment detector with
effector choice
![Page 20: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/20.jpg)
Summary
What makes a system suitable for ARTIS? Pattern classification and Response Distributed architecture, scalable to arbitrary
number of nodes Require detection of novel anomalous
patterns Dynamic but normal behavior changes slowly Robust solution with no central control
![Page 21: Architecture of an Artificial Immune System02317/slides/lec_17.pdf · Architecture of an Artificial Immune System Steven A. Hofmeyr, S. Forrest Presentation by . Ranjani Srinivasan](https://reader034.fdocuments.in/reader034/viewer/2022042404/5f1b781888cdae1602031c9e/html5/thumbnails/21.jpg)
Criticism
Activation Threshold reduces false positive. But introduces paths of attack - Infrequent anomalous connections - Attack has fewer connections than threshold
Assumption of infrequently occurring non-self LRU model assumption