Abstract not found

In previous work, we have introduced a network-based model that abstracts many details of the underlying landscape and compresses the landscape information into a weighted, oriented graph which we call the local optima network. The vertices of this graph are the local optima of the given fitness landscape, while the arcs are transition probabilities between local optima basins. Here, we extend this formalism to neutral fitness landscapes, which are common in difficult combinatorial search spaces. By using two known neutral variants of the NK family (i.e. NKp and NKq) in which the amount of neutrality can be tuned by a parameter, we show that our new definitions of the optima networks and the associated basins are consistent with the previous definitions for the non-neutral case. Moreover, our empirical study and statistical analysis show that the features of neutral landscapes interpolate smoothly between landscapes with maximum neutrality and non-neutral ones. We found some unknown structural differences between the two studied families of neutral landscapes. But overall, the network features studied confirmed that neutrality, in landscapes with percolating neutral networks, may enhance heuristic search. Our current methodology requires the exhaustive enumeration of the underlying search space. Therefore, sampling techniques should be developed before this analysis can have practical implications. We argue, however, that the proposed model offers a new perspective into the problem difficulty of combinatorial optimization problems and may inspire the design of more effective search heuristics.

Abstract not found

Kimura’s neutral theory of evolution has inspired researchers from the evolutionary computation community to incorporate neutrality into Evolutionary Algorithms (EAs) in the hope that it can aid evolution. The effects of neutrality on evolutionary search have been considered in a number of studies, the results of which, however, have been highly contradictory. In this paper, we analyse the reasons for this and we make an effort to shed some light on neutrality by addressing them. We consider two very simple forms of neutrality: constant neutrality — a neutral network of constant fitness, identically distributed in the whole search space — and bit-wise neutrality, where each phenotypic bit is obtained by transforming a group of genotypic bits via an encoding function. We study these forms of neutrality both theoretically and empirically (both for standard benchmark functions and a class of random MAX-SAT problems) to see how and why they influence the behaviour and performance of a mutation-based EA. In particular, we analyse how the fitness distance correlation of landscapes changes under the effect of different neutral encodings and how phenotypic mutation rates vary as a function of genotypic mutation rates. Both help explain why the behaviour of a mutation-based EA may change so radically as problem, form of neutrality and mutation rate are varied.

Abstract: In this work, we explore and study the implication of having more than one output on a Genetic Programming (GP) graph-representation. This approach, called, Multiple Interactive Outputs in a Single Tree (MIOST) is based on two ideas: (a) Firstly, we defined an approach, called Interactivity Within an Individual (IWI), which is based on a graph-GP representation. Secondly, we add to the individuals created with the IWI approach multiple outputs in their structures and as a result of this, we have MIOST. As first step, we analyse the effects of IWI by using only mutations and analyse its implications (i.e., presence of neutrality). Then, we continue testing the effectiveness of IWI by allowing mutations and the standard GP crossover in the evolutionary process. Finally, we tested the effectiveness of MIOST by using mutations and crossover and conduct extensive empirical results on different evolvable problems of different complexity taken from the literature. The results reported in this work, indicate that the proposed approach has a better overall performance in terms of consistency reaching feasible solutions.

Abstract—Research on semantics in Genetic Programming (GP) has increased over the last number of years. Results in this area clearly indicate that its use in GP considerably increases performance. Many of these semantic-based approaches rely on a trial-and-error method that attempts to find offspring that are semantically different from their parents over a number of trials using the crossover operator (crossover-semantics based- CSB). This, in consequence, has a major drawback: these methods could evaluate thousands of nodes, resulting in paying a high computational cost, while attempting to improve performance by promoting semantic diversity. In this work, we propose a simple and computationally inexpensive method, named semantics in selection, that eliminates the computational cost observed in CSB approaches. We tested this approach in 14 GP problems, including continuous- and discrete-valued fitness functions, and compared it against a traditional GP and a CSB approach. Our results are equivalent, and in some cases, superior than those found by the CSB approach, without the necessity of using a “brute force ” mechanism. I.

A key change in the move to Smart Grids (SGs) is the use of dynamic pricing; this together with less reliable energy from renewable resources makes optimising electricity use highly complex. For smart-devices to function in this envi-ronment, they must adapt to this complexity, while main-taining the flexibility to handle changing user behaviour pat-terns. Reinforcement Learning (RL) has been used to op-timise the scheduling of dynamic resources in SGs. It is proposed to provide smart-devices with knowledge of user intentions and actions by leveraging participatory sensing data. This, in consequence, will allow devices in the SG to tailor their operational schedule to users ’ behaviour. With-out this data, the devices ’ operation would be interrupted by user activity, leading to suboptimal results. Participa-tory sensing provides for both, the monitoring of parame-ters affecting devices operation (for example, temperature for a heating system) and access to detailed information about user behaviour and activity. The results obtained by our RL approach, clearly indicate that participatory sensing data indeed improve the performance of device scheduling when compared to static schemes resulting in a dramatic price reduction.

Abstract. Whether neutrality has positive or negative effects on evolutionary search is a contentious topic, with reported experimental results supporting both sides of the debate. Most existing studies use performance statistics, e.g., success rate or search efficiency, to investigate if neutrality, either embedded or artificially added, can benefit an evolutionary algorithm. Here, we argue that understanding the influence of neutrality on evolutionary optimization requires an understanding of the interplay between robustness and evolvability at the genotypic and phenotypic scales. As a concrete example, we consider a simple linear genetic programming system that is amenable to exhaustive enumeration, and allows for the full characterization of these properties. We adopt statistical measurements from RNA systems to quantify robustness and evolvability at both genotypic and phenotypic levels. Using an ensemble of random walks, we demonstrate that the benefit of neutrality crucially depends upon its phenotypic distribution. 1

We present a new perspective of search in evolutionary computing (EC) by using a novel model for the analysis and visualization of genotype to phenotype maps. The model groups genes into quotient sets and shows their adjacencies. A unique quality of the quotient model is that it details geometric qualities of maps that are not otherwise easy to observe. The model shows how random mutations on genes make non-random phenotype preferences, based on the structure of a map. The interaction between such mutation-based preferences with fitness preferences is important for explaining population movements on neutral landscapes. We show the widespread applicability of our approach by applying it to different representations, encodings, and problems including grammatical evolution (GE), Cartesian genetic programming, parity and majority coding, OneMax, Needle-in-Haystack, deceptive trap and hierarchical if-and-only-if. We also use the approach to address conflicting results in the neutral evolution literature and to analyze concepts relevant to neutral evolution including robustness, evolvability, tunneling, and the relation between genetic form and function. We use the model to develop theoretical results on how mapping and neutral evolution affect search in GE. We study the two phases of mapping in GE, these being transcription (i.e., unique identification of genes with integers) and translation (i.e., many-to-one mapping of genotypes to phenotypes). It is shown that translation and transcription schemes belong to equivalence classes, and therefore the properties we derive for specific schemes are applicable to classes of schemes. We present a new perspective on population diversity. We specify conditions under which increasing degeneracy (by increasing codon size) or rearranging the rules of a grammar do not affect performance. It is shown that there is a barrier to nontrivial neutral evolution with the use of the natural transcription with modulo translation combination; a necessary but not sufficient condition for such evolution is that at least three bits should change on mutation within a single codon. This barrier can be avoided by using Gray transcription. We empirically validate some findings.

Abstract not found