Using a map-based encoding to evolve plastic neural networks

Tonelli, Paul; Mouret, Jean-Baptiste

doi:10.1109/eais.2011.5945909

Cited by 6 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, if a genotype for a neutral network is evolved first through training, a plastic neural network would be one that is able to learn and respond to the environment it is placed in, to alter the final weights of the network, leading to a plastic phenotype after genetic evolution. Plastic neural networks have been explored in the past [ 77 – 79 ] with success, showing the benefits of including phenotypic plasticity in intelligent systems that need to run and adapt in the environments they are placed in. Learning can be further extended as a developmental process to include altering the topology of the network, although this has yet to be explored.…”

Section: Missing Concepts From the Extended Evolutionary Synthesismentioning

confidence: 99%

Epigenetic opportunities for evolutionary computation

2023

View full text Add to dashboard Cite

Evolutionary computation is a group of biologically inspired algorithms used to solve complex optimization problems. It can be split into evolutionary algorithms, which take inspiration from genetic inheritance, and swarm intelligence algorithms, that take inspiration from cultural inheritance. However, much of the modern evolutionary literature remains relatively unexplored. To understand which evolutionary mechanisms have been considered, and which have been overlooked, this paper breaks down successful bioinspired algorithms under a contemporary biological framework based on the extended evolutionary synthesis, an extension of the classical, genetics focused, modern synthesis. Although the idea of the extended evolutionary synthesis has not been fully accepted in evolutionary theory, it presents many interesting concepts that could provide benefits to evolutionary computation. The analysis shows that Darwinism and the modern synthesis have been incorporated into evolutionary computation but the extended evolutionary synthesis has been broadly ignored beyond: cultural inheritance, incorporated in the sub-set of swarm intelligence algorithms, evolvability, through covariance matrix adaptation evolution strategy (CMA-ES), and multilevel selection, through multilevel selection genetic algorithm (MLSGA). The framework shows a gap in epigenetic inheritance for evolutionary computation, despite being a key building block in modern interpretations of evolution. This leaves a diverse range of biologically inspired mechanisms as low hanging fruit that should be explored further within evolutionary computation and illustrates the potential of epigenetic based approaches through the recent benchmarks in the literature.

show abstract

Section: Missing Concepts From the Extended Evolutionary Synthesismentioning

confidence: 99%

Epigenetic opportunities for evolutionary computation

2023

View full text Add to dashboard Cite

show abstract

“…Possibly one of the most difficult tasks studied to date, in terms of the size of the exploration space and the number of associations to memorise, is a purely associative task in which agents learn associations between each possible input and output pattern [33,34]. Aside from an input reward signal, the input and output vectors were four binary values.…”

Section: Previous Workmentioning

confidence: 99%

Automated generation of environments to test the general learning capabilities of AI agents

Coleman

Blair

Clune

2014

Proceedings of the 2014 Annual Conference on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

Algorithms for evolving agents that learn during their lifetime have typically been evaluated on only a handful of environments. Designing such environments is labour intensive, potentially biased, and provides only a small sample size that may prevent accurate general conclusions from being drawn. In this paper we introduce a method for automatically generating MDP environments which allows the difficulty to be scaled in several ways. We present a case study in which environments are generated that vary along three key dimensions of difficulty: the number of environment configurations, the number of available actions, and the length of each trial. The study reveals interesting differences between three neural network models -Fixed-Weight, Plastic-Weight, and Modulated Plasticity -that would not have been obvious without sweeping across these different dimensions. Our paper thus introduces a new way of conducting reinforcement learning science: instead of manually designing a few environments, researchers will be able to automatically generate a range of environments across key dimensions of variation. This will allow scientists to more rigorously assess the general learning capabilities of an algorithm, and may ultimately improve the rate at which we discover how to create AI with general purpose learning.

show abstract

“…One of the first parameterised rules was introduced by Niv et al in 2002 [20] and has been used by several subsequent authors [22,23,25,26,28,29]. It consists of a correlation term, pre-and post-synaptic terms and a constant term for heterosynaptic updates, with evolved coefficients for each term:…”

Section: Synaptic Plasticity Modelsmentioning

confidence: 99%

“…Tonelli and Mouret [28,29] studied a purely associative task, similar to those using a supervised learning paradigm, in which an agent is required to learn associations between each possible input and output pattern. The input and output were vectors of binary values, with a reward signal added to the input.…”

Section: Problem Domains Tasks and Fitness Functionsmentioning

confidence: 99%

Evolving Plastic Neural Networks for Online Learning: Review and Future Directions

Coleman

Blair

2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Recent years have seen a resurgence of interest in evolving plastic neural networks for online learning. These approaches have an intrinsic appeal -since, to date, the only working example of general intelligence is the human brain, which has developed through evolution, and exhibits a great capacity to adapt to unfamiliar environments. In this paper we review prior work in this area -including problem domains and tasks, fitness functions, synaptic plasticity models and neural network encoding schemes. We conclude with a discussion of current findings and promising future directions, including incorporation of functional properties observed in biological neural networks which appear to play a role in learning processes, and addressing the "general" in general intelligence by the introduction of previously unseen tasks during the evolution process.

show abstract

Using a map-based encoding to evolve plastic neural networks

Cited by 6 publications

References 26 publications

Epigenetic opportunities for evolutionary computation

Epigenetic opportunities for evolutionary computation

Automated generation of environments to test the general learning capabilities of AI agents

Evolving Plastic Neural Networks for Online Learning: Review and Future Directions

Contact Info

Product

Resources

About