Skip Nav Destination
Close Modal
Update search
NARROW
Date
Availability
1-4 of 4
Extended Papers Invited from GECCO 2017 (Berlin)
Close
Follow your search
Access your saved searches in your account
Would you like to receive an alert when new items match your search?
Sort by
Journal Articles
Publisher: Journals Gateway
Evolutionary Computation (2018) 26 (3): 441–469.
Published: 01 September 2018
FIGURES
| View All (7)
Abstract
View article
PDF
Conventional genetic programming (GP) can guarantee only that synthesized programs pass tests given by the provided input-output examples. The alternative to such a test-based approach is synthesizing programs by formal specification, typically realized with exact, nonheuristic algorithms. In this article, we build on our earlier study on Counterexample-Based Genetic Programming (CDGP), an evolutionary heuristic that synthesizes programs from formal specifications. The candidate programs in CDGP undergo formal verification with a Satisfiability Modulo Theory (SMT) solver, which results in counterexamples that are subsequently turned into tests and used to calculate fitness. The original CDGP is extended here with a fitness threshold parameter that decides which programs should be verified, a more rigorous mechanism for turning counterexamples into tests, and other conceptual and technical improvements. We apply it to 24 benchmarks representing two domains: the linear integer arithmetic (LIA) and the string manipulation (SLIA) problems, showing that CDGP can reliably synthesize provably correct programs in both domains. We also confront it with two state-of-the art exact program synthesis methods and demonstrate that CDGP effectively trades longer synthesis time for smaller program size.
Journal Articles
Publisher: Journals Gateway
Evolutionary Computation (2018) 26 (3): 347–380.
Published: 01 September 2018
FIGURES
| View All (9)
Abstract
View article
PDF
Algorithms that learn through environmental interaction and delayed rewards, or reinforcement learning (RL), increasingly face the challenge of scaling to dynamic, high-dimensional, and partially observable environments. Significant attention is being paid to frameworks from deep learning, which scale to high-dimensional data by decomposing the task through multilayered neural networks. While effective, the representation is complex and computationally demanding. In this work, we propose a framework based on genetic programming which adaptively complexifies policies through interaction with the task. We make a direct comparison with several deep reinforcement learning frameworks in the challenging Atari video game environment as well as more traditional reinforcement learning frameworks based on a priori engineered features. Results indicate that the proposed approach matches the quality of deep learning while being a minimum of three orders of magnitude simpler with respect to model complexity. This results in real-time operation of the champion RL agent without recourse to specialized hardware support. Moreover, the approach is capable of evolving solutions to multiple game titles simultaneously with no additional computational cost. In this case, agent behaviours for an individual game as well as single agents capable of playing all games emerge from the same evolutionary run.
Journal Articles
Publisher: Journals Gateway
Evolutionary Computation (2018) 26 (3): 381–410.
Published: 01 September 2018
FIGURES
| View All (16)
Abstract
View article
PDF
Design optimization techniques are often used at the beginning of the design process to explore the space of possible designs. In these domains illumination algorithms, such as MAP-Elites, are promising alternatives to classic optimization algorithms because they produce diverse, high-quality solutions in a single run, instead of only a single near-optimal solution. Unfortunately, these algorithms currently require a large number of function evaluations, limiting their applicability. In this article, we introduce a new illumination algorithm, Surrogate-Assisted Illumination (SAIL), that leverages surrogate modeling techniques to create a map of the design space according to user-defined features while minimizing the number of fitness evaluations. On a two-dimensional airfoil optimization problem, SAIL produces hundreds of diverse but high-performing designs with several orders of magnitude fewer evaluations than MAP-Elites or CMA-ES. We demonstrate that SAIL is also capable of producing maps of high-performing designs in realistic three-dimensional aerodynamic tasks with an accurate flow simulation. Data-efficient design exploration with SAIL can help designers understand what is possible, beyond what is optimal, by considering more than pure objective-based optimization.
Journal Articles
Publisher: Journals Gateway
Evolutionary Computation (2018) 26 (3): 411–440.
Published: 01 September 2018
FIGURES
| View All (32)
Abstract
View article
PDF
The hypervolume indicator has frequently been used for comparing evolutionary multi-objective optimization (EMO) algorithms. A reference point is needed for hypervolume calculation. However, its specification has not been discussed in detail from a viewpoint of fair performance comparison. A slightly worse point than the nadir point is usually used for hypervolume calculation in the EMO community. In this paper, we propose a reference point specification method for fair performance comparison of EMO algorithms. First, we discuss the relation between the reference point specification and the optimal distribution of solutions for hypervolume maximization. It is demonstrated that the optimal distribution of solutions strongly depends on the location of the reference point when a multi-objective problem has an inverted triangular Pareto front. Next, we propose a reference point specification method based on theoretical discussions on the optimal distribution of solutions. The basic idea is to specify the reference point so that a set of well-distributed solutions over the entire linear Pareto front has a large hypervolume and all solutions in such a solution set have similar hypervolume contributions. Then, we examine whether the proposed method can appropriately specify the reference point through computational experiments on various test problems. Finally, we examine the usefulness of the proposed method in a hypervolume-based EMO algorithm. Our discussions and experimental results clearly show that a slightly worse point than the nadir point is not always appropriate for performance comparison of EMO algorithms.