On the Use of Hyperspheres in Artificial Immune Systems as Antibody Recognition Regions

Stibor, Thomas; Timmis, Jon; Eckert, Claudia

doi:10.1007/11823940_17

Cited by 36 publications

(18 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The spherical case has already been discussed by Stibor et al in the context of negative selection algorithms [55]. The reason for our simplification is two-fold: firstly, in some respects the overall message is diluted by the mathematics of approximating the volume of a hyper-sphere and the particularly bizarre result that follows -as dimensionality increases, the volume of the hyper-sphere approaches zero.…”

Section: Theoretical Issuesmentioning

confidence: 97%

Representation in the (Artificial) Immune System

McEwan

Hart

2009

J Math Model Algor

View full text Add to dashboard Cite

Abstract. Much of contemporary research in Artificial Immune Systems (AIS) has partitioned into either algorithmic machine learning and optimisation, or, modelling biologically plausible dynamical systems, with little overlap between. We propose that this dichotomy is somewhat to blame for the lack of significant advancement of the field in either direction and demonstrate how a simplistic interpretation of Perelson's shape-space formalism may have largely contributed to this dichotomy. In this paper, we motivate and derive an alternative representational abstraction. To do so we consider the validity of shape-space from both the biological and machine learning perspectives. We then take steps towards formally integrating these perspectives into a coherent computational model of notions such as life-long learning, degeneracy, constructive representations and contextual recognition -rhetoric that has long inspired work in AIS, while remaining largely devoid of operational definition.

show abstract

Section: Theoretical Issuesmentioning

confidence: 97%

Representation in the (Artificial) Immune System

McEwan

Hart

2009

J Math Model Algor

View full text Add to dashboard Cite

show abstract

“…Although the key concept of mapping immunological shape space to a vectorial representation of a data-set has driven much of successful AIS research, recent theoretical investigations have cast doubt on the tractability of the concept (see McEwan and Hart 2009;Stibor et al 2006). Instance-based methods rely on two key assumptions; that there are dense regions in the space that can be generalised, compressed or sparsely represented, and that the distance between points is a meaningful proxy for comparison, discrimination and localisation.…”

Section: Immune Swarms For Engineeringmentioning

confidence: 99%

On artificial immune systems and swarm intelligence

2010

View full text Add to dashboard Cite

This position paper explores the nature and role of two bio-inspired paradigms, namely Artificial Immune Systems (AIS) and Swarm Intelligence (SI). We argue that there are many aspects of AIS that have direct parallels with SI and examine the role of AIS and SI in science and also in engineering, with the primary focus being on the immune system. We explore how in some ways, algorithms from each area are similar, but we also advocate, and explain, that rather than being competitors, AIS and SI are complementary tools and can be used effectively together to solve complex engineering problems.

show abstract

“…Many previous works have been studied rs in detail [7] [15] [18], so in this work it is not discussed. According to Eq.12 this paper calculated the minimum distant (d min ) between self sample and nonself sample on Haberman's Survival Data Set and KDDCUP99.…”

Section: Parameters Setting (1)the Radius Of Self Sample (Rs)mentioning

confidence: 99%

A real negative selection algorithm with evolutionary preference for anomaly detection

Yang¹,

Wen

2017

Open Physics

View full text Add to dashboard Cite

Traditional real negative selection algorithms (RNSAs) adopt the estimated coverage (c 0 ) as the algorithm termination threshold, and generate detectors randomly. With increasing dimensions, the data samples could reside in the low-dimensional subspace, so that the traditional detectors cannot effectively distinguish these samples. Furthermore, in high-dimensional feature space, c 0 cannot exactly reflect the detectors set coverage rate for the nonself space, and it could lead the algorithm to be terminated unexpectedly when the number of detectors is insufficient. These shortcomings make the traditional RNSAs to perform poorly in high-dimensional feature space. Based upon "evolutionary preference" theory in immunology, this paper presents a real negative selection algorithm with evolutionary preference (RNSAP). RNSAP utilizes the "unknown nonself space", "low-dimensional target subspace" and "known nonself feature" as the evolutionary preference to guide the generation of detectors, thus ensuring the detectors can cover the nonself space more effectively. Besides, RNSAP uses redundancy to replace c 0 as the termination threshold, in this way RNSAP can generate adequate detectors under a proper convergence rate. The theoretical analysis and experimental result demonstrate that, compared to the classical RNSA (V-detector), RNSAP can achieve a higher detection rate, but with less detectors and computing cost.

show abstract

On the Use of Hyperspheres in Artificial Immune Systems as Antibody Recognition Regions

Cited by 36 publications

References 17 publications

Representation in the (Artificial) Immune System

Representation in the (Artificial) Immune System

On artificial immune systems and swarm intelligence

A real negative selection algorithm with evolutionary preference for anomaly detection

Contact Info

Product

Resources

About