Mario Pavone

Exploring the Capability of Immune Algorithms: A Characterization of Hypermutation Operators

In this paper, an important class of hypermutation operators are discussed and quantitatively compared with respect to their success rate and computational cost. We use a standard Immune Algorithm (IA), based on the clonal selection principle to investigate the searching capability of the designed hypermutation operators. We computed the parameter surface for each variation operator to predict the best parameter setting for each operator and their combination. The experimental investigation in which we use a standard clonal selection algorithm with different hypermutation operators on a complex “toy problem”, the trap functions, and a complex NP-complete problem, the 2D HP model for the protein structure prediction problem, clarifies that only few really different and useful hypermutation operators exist, namely: inversely proportional hypermutation, static hypermutation and hypermacromutation operators. The combination of static and inversely proportional Hypermutation and hypermacromutation showed the best experimental results for the “toy problem” and the NP-complete problem.

Clonal selection algorithms: a comparative case study using effective mutation potentials

This paper presents a comparative study of two important Clonal Selection Algorithms (CSAs): CLONALG and opt-IA. To deeply understand the performance of both algorithms, we deal with four different classes of problems: toy problems (one-counting and trap functions), pattern recognition, numerical optimization problems and NP-complete problem (the 2D HP model for protein structure prediction problem). Two possible versions of CLONALG have been implemented and tested. The experimental results show a global better performance of opt-IA with respect to CLONALG. Considering the results obtained, we can claim that CSAs represent a new class of Evolutionary Algorithms for effectively performing searching, learning and optimization tasks.

Parallel Problem Solving from Nature - PPSN XII

The information geometric optimization (IGO) flow has been introduced recently by Arnold et al. This distinguished mathematical flow on the parameter manifold of a family of search distributions constitutes a novel approach to the analysis of several randomized search heuristics, including modern evolution strategies. Besides its appealing theoretical properties, it offers the unique opportunity to approach the convergence analysis of evolution strategies in two independent steps. The first step is the analysis of the flow itself, or more precisely, the convergence of its trajectories to Dirac peaks over the optimum. In a second step it remains to study the deviation of actual algorithm trajectories from the continuous flow. The present study approaches the first problem. The IGO flow of isotropic Gaussian search distributions is analyzed on convex, quadratic fitness functions. Convergence of all trajectories to the Dirac peak over the optimum is established.

A hybrid immune algorithm with information gain for the graph coloring problem

We present a new Immune Algorithm that incorporates a simple local search procedure to improve the overall performances to tackle the graph coloring problem instances. We characterize the algorithm and set its parameters in terms of Information Gain. Experiments will show that the IA we propose is very competitive with the best evolutionary algorithms.

Real coded clonal selection algorithm for unconstrained global optimization using a hybrid inversely proportional hypermutation operator

Numerical optimization of given objective functions is a crucial task in many real-life problems. This paper introduces a new immunological algorithm for continuous global optimization problems, called opt-IMMALG; it is an improved version of a previously proposed clonal selection algorithm, using a real-code representation and a new Inversely Proportional Hypermutation operator.We evaluate and assess the performance of opt-IMMALG and several others algorithms, namely opt-IA, PSO, arPSO, DE, and SEA with respect to their general applicability as numerical optimization algorithms. The experiments have been performed on 23 widely used benchmark problems.The experimental results show that opt-IMMALG is a suitable numerical optimization technique that, in terms of accuracy, outperforms the analyzed algorithms in this comparative study. In addition it is shown that opt-IMMALG is also suitable for solving large-scale problems.

Immune algorithms with aging operators for the string folding problem and the protein folding problem

We present an Immune Algorithm (IA) based on clonal selection principle and which uses memory B cells, to face the protein structure prediction problem (PSP) a particular example of the String Folding Problem in 2D and 3D lattice. Memory B cells with a longer life span are used to partition the funnel landscape of PSP, so to properly explore the search space. The designed IA shows its ability to tackle standard benchmarks instances substantially better than other IA’s. In particular, for the 3D HP model the IA allowed us to find energy minima not found by other evolutionary algorithms described in literature.

An immune algorithm with stochastic aging and kullback entropy for the chromatic number problem

We present a new Immune Algorithm, IMMALG, that incorporates a Stochastic Aging operator and a simple local search procedure to improve the overall performances in tackling the chromatic number problem (CNP) instances. We characterize the algorithm and set its parameters in terms of Kullback Entropy. Experiments will show that the IA we propose is very competitive with the state-of-art evolutionary algorithms.

An immune algorithm with hyper-macromutations for the Dill's 2D hydrophobic-hydrophilic model

This paper presents an immune algorithm (IA) based on clonal selection principle using a mutation operator, the hypermacromutation, and an aging process to tackle the protein structure prediction problem (PSP) in the 2D hydrophobic-hydrophilic model. The IA presented has only three parameters. To correctly set these parameters we compute the parameter surfaces, the 3D plots of IA success rate in function of the cloning parameter and the maximum age allowed to each B cell. The parameter surfaces show that hypermacromutation and aging operators are key features for generating diversity and searching more properly the funnel landscape of the PSP problem. Experiments show that the immune algorithm we propose is very competitive with the state-of-art algorithms for the PSP.

An immunological algorithm for global numerical optimization

Numerical optimization of given objective functions is a crucial task in many real-life problems. The present article introduces an immunological algorithm for continuous global optimization problems, called OPT-IA. Several biologically inspired algorithms have been designed during the last few years and have shown to have very good performance on standard test bed for numerical optimization. In this paper we assess and evaluate the performance of OPT-IA, FEP, IFEP, DIRECT, CEP, PSO, and EO with respect to their general applicability as numerical optimization algorithms. The experimental protocol has been performed on a suite of 23 widely used benchmarks problems. The experimental results show that OPT-IA is a suitable numerical optimization technique that, in terms of accuracy, generally outperforms the other algorithms analyzed in this comparative study. The OPT-IA is also shown to be able to solve large-scale problems.

Protein multiple sequence alignment by hybrid bio-inspired algorithms

This article presents an immune inspired algorithm to tackle the Multiple Sequence Alignment (MSA) problem. MSA is one of the most important tasks in biological sequence analysis. Although this paper focuses on protein alignments, most of the discussion and methodology may also be applied to DNA alignments. The problem of finding the multiple alignment was investigated in the study by Bonizzoni and Vedova and Wang and Jiang, and proved to be a NP-hard (non-deterministic polynomial-time hard) problem. The presented algorithm, called Immunological Multiple Sequence Alignment Algorithm (IMSA), incorporates two new strategies to create the initial population and specific ad hoc mutation operators. It is based on the ‘weighted sum of pairs’ as objective function, to evaluate a given candidate alignment. IMSA was tested using both classical benchmarks of BAliBASE (versions 1.0, 2.0 and 3.0), and experimental results indicate that it is comparable with state-of-the-art multiple alignment algorithms, in terms of quality of alignments, weighted Sums-of-Pairs (SP) and Column Score (CS) values. The main novelty of IMSA is its ability to generate more than a single suboptimal alignment, for every MSA instance; this behaviour is due to the stochastic nature of the algorithm and of the populations evolved during the convergence process. This feature will help the decision maker to assess and select a biologically relevant multiple sequence alignment. Finally, the designed algorithm can be used as a local search procedure to properly explore promising alignments of the search space.