W.B.Langdon . 21 August 2014 2005 papers , full list
ABSTRACT
We report initial experiments on evolving kernel functions which describe the average behaviour of a swarm of particles as if it was responding as a single point moving on a landscape transformed by the kernel. Such kernels may help explain swarm and population approaches leading to extended population systems, XPS. The swarm of particles are from a simple particle swarm optimiser solving one dimensional multi-modal 3 peaks and Rastrigin problems. The standard Java genetic programming implementation TinyGP is used.
ABSTRACT
An Internet Java Applet allows users anywhere to play the Solo Pong game. We compare people's performance to a hand coded ``Optimal'' player and programs automatically produced by artificial intelligence. The AI techniques are: genetic programming, including a hybrid of GP and a human designed algorithm, and a particle swarm optimiser. The AI approaches are not fine tuned. GP and PSO find good players. Evolutionary computation (EC) is able to beat both human designed code and human players.
Two page version (PDF ps.gz) presented at BNAIC 2005 pp 365-366.
ABSTRACT
We use evolutionary computation (EC) to automatically find problems which demonstrate the strength and weaknesses of modern search heuristics. In particular we analyse Particle Swarm Optimization (PSO) and Differential Evolution (DE). Both evolutionary algorithms are contrasted with a robust deterministic gradient based searcher (based on Newton-Raphson). The fitness landscapes made by genetic programming (GP) are used to illustrate difficulties in GAs and PSOs thereby explaining how they work and allowing us to devise better extended particle swarm systems (XPS).
ABSTRACT
Fitness distributions (landscapes) of programs tend to a limit as they get bigger. Markov minorization gives upper bounds ((15.3 + 2.30 m)/log(I)) on the length of program run on random or average computing devices. I is the size of the instruction set and m size of output register. Almost all programs are constants. Convergence is exponential with 90% of programs of length 1.6 n 2**N yielding constants (n=size input register and size of memory=N). This is supported by experiment.
ABSTRACT
An Internet Java Applet allows users anywhere to play the Solo Pong game. We compare people's performance to a hand coded ``Optimal'' player and programs automatically produced by artificial intelligence. The AI techniques are: genetic programming, including a hybrid of GP and a human designed algorithm, and a particle swarm optimiser. The AI approaches are not fine tuned. GP and PSO find good players. Evolutionary computation (EC) is able to beat both human designed code and human players.
Two animated slides: constriction helps, hinders.
ABSTRACT
Genetic programming (GP) is used to create fitness landscapes which highlight strengths and weaknesses of different types of PSO and to contrast population-based swarm approaches with non stochastic gradient followers (i.e. hill climbers). These automatically generated benchmark problems yield insights into the operation of PSOs, illustrate benefits and drawbacks of different population sizes and constriction (friction) coefficients, and reveal new swarm phenomena such as deception and the exploration/exploitation tradeoff. The method could be applied to any type of optimizer.
ABSTRACT
Pfeiffer contains a population of fractals which has been evolving continuously for more than three years. The animations are developed from embryos using a Lindenmayer grammar (L-System). These open generative representations potentially allow gene duplication and the evolution of higher order genetic operators and might be a step towards the emergence of social intelligence in swarms of artificial life (alife) agents. The fitness function is simply do the snowflake patterns appeal to the users: interactive evolution (IEC). To this end, images are placed in animated snow globes (computerised snowstorms) by world wide web (www) browsers (Netscape, Mozilla, Internet Explorer, Firefox, etc.) anywhere on the planet. More than 600 people have used http://www.cs.ucl.ac.uk/staff/W.Langdon/pfeiffer.html.
ABSTRACT
We extend our analysis of repetitive patterns found in genetic programming genomes * to tree based GP. As in linear GP, repetitive patterns are present in large numbers. Size fair crossover limits bloat in automatic programming, preventing the evolution of recurring motifs. We examine these complex properties in detail: e.g. using depth v. size Catalan binary tree shape plots, subgraph and subtree matching, information entropy, syntactic and semantic fitness correlations and diffuse introns. We relate this emergent phenomenon to considerations about building blocks in GP and how GP works.
ABSTRACT
The journal, and in particular the resource reviews have been running for more than five years. Now is a good time to revisit our original goals compare them with what the journal has achieved and make new plans. Section 2 onwards, updates the statistics on the state of the genetic programming, evolvable hardware and evolvable machines literature and electronic resources.
ABSTRACT
Biological chromosomes are replete with repetitive sequences, microsatellites, SSR tracts, ALU, etc. in their DNA base sequences. We started looking for similar phenomena in evolutionary computation. First studies find copious repeated sequences, which can be hierarchically decomposed into shorter sequences, in programs evolved using both homologous and two point crossover but not with headless chicken crossover or other mutations. In bloated programs the small number of effective or expressed instructions appear in both repeated and non-repeated code. Hinting that building-blocks or code reuse may evolve in unplanned ways.
Mackey-Glass chaotic time series prediction and eukaryotic protein localisation (both previously used as artificial intelligence machine learning benchmarks) demonstrate evolution of Shannon information (entropy) and lead to models capable of lossy Kolmogorov compression. Our findings with diverse benchmarks and GP systems suggest this emergent phenomenon may be widespread in genetic systems.
Data set Bibliographic details