Using novelty search to explicitly create diversity in ensembles of classifiers
Conference Proceeding
Cardoso, R. P., Hart, E., Kurka, D. B., & Pitt, J. V. (2021)
Using novelty search to explicitly create diversity in ensembles of classifiers. In GECCO '21: Proceedings of the Genetic and Evolutionary Computation Conference (849-857). https://doi.org/10.1145/3449639.3459308
The diversity between individual learners in an ensemble is known to influence its performance. However, there is no standard agreement on how diversity should be defined, and...
Automated, Explainable Rule Extraction from MAP-Elites archives
Conference Proceeding
Urquhart, N., Höhl, S., & Hart, E. (2021)
Automated, Explainable Rule Extraction from MAP-Elites archives. In Applications of Evolutionary Computation: 24th International Conference, EvoApplications 2021. , (258-272). https://doi.org/10.1007/978-3-030-72699-7_17
Quality-diversity(QD) algorithms that return a large archive of elite solutions to a problem provide insights into how high-performing solutions are distributed throughout a f...
WILDA: Wide Learning of Diverse Architectures for Classification of Large Datasets
Conference Proceeding
Pitt, J., Burth Kurka, D., Hart, E., & Cardoso, R. P. (2021)
WILDA: Wide Learning of Diverse Architectures for Classification of Large Datasets. In Applications of Evolutionary Computation: EvoApplications 2021 Proceedings. , (649-664). https://doi.org/10.1007/978-3-030-72699-7_41
In order to address scalability issues, which can be a challenge for Deep Learning methods, we propose Wide Learning of Diverse Architectures-a model that scales horizontally ...
Generating Unambiguous and Diverse Referring Expressions
Journal Article
Panagiaris, N., Hart, E., & Gkatzia, D. (2021)
Generating Unambiguous and Diverse Referring Expressions . Computer Speech and Language, 68, Article 101184. https://doi.org/10.1016/j.csl.2020.101184
Neural Referring Expression Generation (REG) models have shown promising results in generating expressions which uniquely describe visual objects. However, current REG models ...
Improving the Naturalness and Diversity of Referring Expression Generation models using Minimum Risk Training
Conference Proceeding
Panagiaris, N., Hart, E., & Gkatzia, D. (2020)
Improving the Naturalness and Diversity of Referring Expression Generation models using Minimum Risk Training. In Proceedings of the 13th International Conference on Natural Language Generation. , (41-51
In this paper we consider the problem of optimizing neural Referring Expression Generation (REG) models with sequence level objectives. Recently reinforcement learning (RL) te...
Evolution of Diverse, Manufacturable Robot Body Plans
Conference Proceeding
Buchanan, E., Le Goff, L., Hart, E., Eiben, A. E., De Carlo, M., Li, W., …Tyrrell, A. M. (2020)
Evolution of Diverse, Manufacturable Robot Body Plans. In 2020 IEEE Symposium Series on Computational Intelligence (SSCI) (2132-2139). https://doi.org/10.1109/SSCI47803.2020.9308434
Advances in rapid prototyping have opened up new avenues of research within Evolutionary Robotics in which not only controllers but also the body plans (morphologies) of robot...
Hardware Design for Autonomous Robot Evolution
Conference Proceeding
Hale, M. F., Angus, M., Buchanan, E., Li, W., Woolley, R., Le Goff, L. K., …Tyrrell, A. M. (2020)
Hardware Design for Autonomous Robot Evolution. In 2020 IEEE Symposium Series on Computational Intelligence (SSCI) (2140-2147). https://doi.org/10.1109/SSCI47803.2020.9308204
The long term goal of the Autonomous Robot Evolution (ARE) project is to create populations of physical robots, in which both the controllers and body plans are evolved. The t...
Bootstrapping artificial evolution to design robots for autonomous fabrication
Journal Article
Buchanan, E., Le Goff, L. K., Li, W., Hart, E., Eiben, A. E., De Carlo, M., …Tyrrell, A. M. (2020)
Bootstrapping artificial evolution to design robots for autonomous fabrication. Robotics, 9(4), https://doi.org/10.3390/robotics9040106
A long-term vision of evolutionary robotics is a technology enabling the evolution of entire autonomous robotic ecosystems that live and work for long periods in challenging a...
Improving Classification of Metamorphic Malware by Augmenting Training Data with a Diverse Set of Evolved Mutant Samples
Conference Proceeding
Babaagba, K., Tan, Z., & Hart, E. (2020)
Improving Classification of Metamorphic Malware by Augmenting Training Data with a Diverse Set of Evolved Mutant Samples. https://doi.org/10.1109/CEC48606.2020.9185668
Detecting metamorphic malware provides a challenge to machine-learning models as trained models might not generalise to future mutant variants of the malware. To address this,...
On Pros and Cons of Evolving Topologies with Novelty Search
Conference Proceeding
Le Goff, L. K., Hart, E., Coninx, A., & Doncieux, S. (2020)
On Pros and Cons of Evolving Topologies with Novelty Search. In ALIFE 2020: The 2020 Conference on Artificial Life. , (423-431). https://doi.org/10.1162/isal_a_00291
Novelty search was proposed as a means of circumventing deception and providing selective pressure towards novel behaviours to provide a path towards open-ended evolution. Ini...