Neuroevolution has re-emerged as an active topic in the last few years. However, there is a lack ... more Neuroevolution has re-emerged as an active topic in the last few years. However, there is a lack of accessible tools to analyse, contrast and visualise the behaviour of neuroevolution systems. A variety of search strategies have been proposed such as Novelty search and Quality-Diversity search, but their impact on the evolutionary dynamics is not well understood. We propose using a data-driven, graph-based model, search trajectory networks (STNs) to analyse, visualise and directly contrast the behaviour of different neuroevolution search methods. Our analysis uses NEAT for solving maze problems with two search strategies: novelty-based and fitness-based, and including and excluding the crossover operator. We model and visualise the trajectories, contrasting and illuminating the behaviour of the studied neuroevolution variants. Our results confirm the advantages of novelty search in this setting, but challenge the usefulness of recombination.
Many interesting phenomena emerge as the result of individual choices made by large numbers of in... more Many interesting phenomena emerge as the result of individual choices made by large numbers of interacting people. To study these phenomena we need to do experiments. But these can be expensive, impractical, or unethical to carry out in the real world. Solution Virtual experiments use simulated scenarios instead of the real world. Participatory simulation means that only small numbers of human participants are needed. Large populations are created using simulated “bots ” who copy the behaviour of the human subjects. Potential applications Study of emergent social phenomena such as: epidemic spread – see example below; stock market behaviour; viral marketing; social networks; spread of rumours and news... Computer gaming back-end simulation of epidemic spreading in a large population Example: application in epidemiology The model shows that the best way to behave in an epidemic is to be very cautious and stay at home until it is over... But what do people really do? front end present...
Many interesting phenomena emerge as the result of individual choices made by large numbers of in... more Many interesting phenomena emerge as the result of individual choices made by large numbers of interacting people. To study these phenomena we need to do experiments. But these can be expensive, impractical, or unethical to carry out in the real world. Solution Virtual experiments use simulated scenarios instead of the real world. Participatory simulation means that only small numbers of human participants are needed. Large populations are created using simulated “bots ” who copy the behaviour of the human subjects. Potential applications Study of emergent social phenomena such as: epidemic spread – see example below; stock market behaviour; viral marketing; social networks; spread of rumours and news... Computer gaming back-end simulation of epidemic spreading in a large population Example: application in epidemiology The model shows that the best way to behave in an epidemic is to be very cautious and stay at home until it is over... But what do people really do? front end present...
Many interesting phenomena emerge as the result of individual choices made by large numbers of in... more Many interesting phenomena emerge as the result of individual choices made by large numbers of interacting people. To study these phenomena we need to do experiments. But these can be expensive, impractical, or unethical to carry out in the real world. Solution Virtual experiments use simulated scenarios instead of the real world. Participatory simulation means that only small numbers of human participants are needed. Large populations are created using simulated “bots ” who copy the behaviour of the human subjects. Potential applications Study of emergent social phenomena such as: epidemic spread – see example below; stock market behaviour; viral marketing; social networks; spread of rumours and news... Computer gaming back-end simulation of epidemic spreading in a large population Example: application in epidemiology The model shows that the best way to behave in an epidemic is to be very cautious and stay at home until it is over... But what do people really do? front end present...
Neuroevolution has re-emerged as an active topic in the last few years. However, there is a lack ... more Neuroevolution has re-emerged as an active topic in the last few years. However, there is a lack of accessible tools to analyse, contrast and visualise the behaviour of neuroevolution systems. A variety of search strategies have been proposed such as Novelty search and Quality-Diversity search, but their impact on the evolutionary dynamics is not well understood. We propose using a data-driven, graph-based model, search trajectory networks (STNs) to analyse, visualise and directly contrast the behaviour of different neuroevolution search methods. Our analysis uses NEAT for solving maze problems with two search strategies: novelty-based and fitness-based, and including and excluding the crossover operator. We model and visualise the trajectories, contrasting and illuminating the behaviour of the studied neuroevolution variants. Our results confirm the advantages of novelty search in this setting, but challenge the usefulness of recombination.
Many interesting phenomena emerge as the result of individual choices made by large numbers of in... more Many interesting phenomena emerge as the result of individual choices made by large numbers of interacting people. To study these phenomena we need to do experiments. But these can be expensive, impractical, or unethical to carry out in the real world. Solution Virtual experiments use simulated scenarios instead of the real world. Participatory simulation means that only small numbers of human participants are needed. Large populations are created using simulated “bots ” who copy the behaviour of the human subjects. Potential applications Study of emergent social phenomena such as: epidemic spread – see example below; stock market behaviour; viral marketing; social networks; spread of rumours and news... Computer gaming back-end simulation of epidemic spreading in a large population Example: application in epidemiology The model shows that the best way to behave in an epidemic is to be very cautious and stay at home until it is over... But what do people really do? front end present...
Many interesting phenomena emerge as the result of individual choices made by large numbers of in... more Many interesting phenomena emerge as the result of individual choices made by large numbers of interacting people. To study these phenomena we need to do experiments. But these can be expensive, impractical, or unethical to carry out in the real world. Solution Virtual experiments use simulated scenarios instead of the real world. Participatory simulation means that only small numbers of human participants are needed. Large populations are created using simulated “bots ” who copy the behaviour of the human subjects. Potential applications Study of emergent social phenomena such as: epidemic spread – see example below; stock market behaviour; viral marketing; social networks; spread of rumours and news... Computer gaming back-end simulation of epidemic spreading in a large population Example: application in epidemiology The model shows that the best way to behave in an epidemic is to be very cautious and stay at home until it is over... But what do people really do? front end present...
Many interesting phenomena emerge as the result of individual choices made by large numbers of in... more Many interesting phenomena emerge as the result of individual choices made by large numbers of interacting people. To study these phenomena we need to do experiments. But these can be expensive, impractical, or unethical to carry out in the real world. Solution Virtual experiments use simulated scenarios instead of the real world. Participatory simulation means that only small numbers of human participants are needed. Large populations are created using simulated “bots ” who copy the behaviour of the human subjects. Potential applications Study of emergent social phenomena such as: epidemic spread – see example below; stock market behaviour; viral marketing; social networks; spread of rumours and news... Computer gaming back-end simulation of epidemic spreading in a large population Example: application in epidemiology The model shows that the best way to behave in an epidemic is to be very cautious and stay at home until it is over... But what do people really do? front end present...
Uploads
Papers by Jason Adair