Theoretical analysis of diversity mechanisms for global exploration

Friedrich, Tobias; Oliveto, Pietro S.; Sudholt, Dirk; Witt, Carsten

doi:10.1145/1389095.1389276

Cited by 37 publications

(45 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditionally it has been believed that crossover may be useful only if sufficient diversity is readily available and that the emergence of diversity in the population is due to either mutation alone or should be enforced by the introduction of diversity mechanisms [4,6,9]. Indeed, previous work highlighting that crossover may be beneficial for Jump k used unrealistically low crossover probabilities to allow mutation alone to create sufficient diversity.…”

Section: Resultsmentioning

confidence: 99%

Emergence of Diversity and Its Benefits for Crossover in Genetic Algorithms

Dang

Friedrich

Kötzing

et al. 2016

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. Population diversity is essential for avoiding premature convergence in Genetic Algorithms (GAs) and for the effective use of crossover. Yet the dynamics of how diversity emerges in populations are not well understood. We use rigorous runtime analysis to gain insight into population dynamics and GA performance for a standard (µ+1) GA and the Jump k test function. By studying the stochastic process underlying the size of the largest collection of identical genotypes we show that the interplay of crossover followed by mutation may serve as a catalyst leading to a sudden burst of diversity. This leads to improvements of the expected optimisation time of order Ω(n/ log n) compared to mutationonly algorithms like the (1+1) EA.

show abstract

Section: Resultsmentioning

confidence: 99%

Emergence of Diversity and Its Benefits for Crossover in Genetic Algorithms

Dang

Friedrich

Kötzing

et al. 2016

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…The effectiveness of fitness sharing has been shown recently (Friedrich, Oliveto, Sudholt, and Witt [4]) for a toy problem that, however, has the same structure as the Mincut instance we will investigate in Section 4.…”

Section: Decrease the Attractiveness Of Local Optimamentioning

confidence: 93%

Memetic algorithms with variable-depth search to overcome local optima

Sudholt

2008

Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation

Self Cite

View full text Add to dashboard Cite

Variable-depth search (shortly VDS) is well-known as LinKernighan strategy for the TSP and Kernighan-Lin for graph partitioning. The basic idea is to make a sequence of local moves and to freeze all moved combinatorial objects to prevent the search from looping. VDS stops when no further local move is possible and returns a best found solution.We analyze memetic algorithms with VDS for three binary combinatorial problems: Mincut, Knapsack, and Maxsat. More precisely, we focus on simply structured problem instances containing local optima that are very hard to overcome. Many common trajectory-based algorithms fail to find a global optimum: the (1+1) EA, iterated local search, and simulated annealing need exponential time with high probability. However, memetic algorithms using VDS easily manage to find a global optimum in expected polynomial time. These results strengthen the usefulness of memetic algorithms with VDS from a theoretical perspective.

show abstract

“…The most popular approach to mitigating premature convergence in evolutionary computation is undoubtedly to foster the diversity of the population [68,159,117,67,42,174,171]. Diversity is typically encouraged in the genotype space, but such an approach is computationally expensive for many genotypes used in evolutionary robotics and, in particular, for neural networks whose topology is evolved [171,131].…”

Section: Behavioral Diversitymentioning

confidence: 99%

Beyond black-box optimization: a review of selective pressures for evolutionary robotics

Doncieux

Mouret

2014

Evol. Intel.

View full text Add to dashboard Cite

Evolutionary robotics is often viewed as the application of a family of black-box optimization algorithms -evolutionary algorithms -to the design of robots, or parts of robots. When considering evolutionary robotics as black-box optimization, the selective pressure is mainly driven by a user-defined, black-box fitness function, and a domain-independent selection procedure. However, most evolutionary robotics experiments face similar challenges in similar setups: the selective pressure, and, in particular, the fitness function, is not a pure user-defined black box. The present review shows that, because evolutionary robotics experiments share common features, selective pressures for evolutionary robotics are a subject of research on their own. The literature has been split into two categories: goal refiners, aimed at changing the definition of a good solution, and process helpers, designed to help the search process. Two subcategories are further considered: task-specific approaches, which require knowledge on how to solve the task and taskagnostic ones, which do not need it. Besides highlighting the diversity of the approaches and their respective goals, the present review shows that many task-agnostic process helpers have been proposed during the last years, thus bringing us closer to the goal of a fully automated robot behavior design process.

show abstract

Theoretical analysis of diversity mechanisms for global exploration

Cited by 37 publications

References 20 publications

Emergence of Diversity and Its Benefits for Crossover in Genetic Algorithms

Emergence of Diversity and Its Benefits for Crossover in Genetic Algorithms

Memetic algorithms with variable-depth search to overcome local optima

Beyond black-box optimization: a review of selective pressures for evolutionary robotics

Contact Info

Product

Resources

About