All Outputs (13)

Quality-Diversity Optimisation on a Physical Robot Through Dynamics-Aware and Reset-Free Learning (2023)
Presentation / Conference Contribution
Smith, S. C., Lim, B., Janmohamed, H., & Cully, A. (2023, July). Quality-Diversity Optimisation on a Physical Robot Through Dynamics-Aware and Reset-Free Learning. Presented at Genetic and Evolutionary Computation Conference Companion (GECCO 2023 Companion), Lisbon

Learning algorithms, like Quality-Diversity (QD), can be used to acquire repertoires of diverse robotics skills. This learning is commonly done via computer simulation due to the large number of evaluations required. However, training in a virtual en... Read More about Quality-Diversity Optimisation on a Physical Robot Through Dynamics-Aware and Reset-Free Learning.

Online Damage Recovery for Physical Robots with Hierarchical Quality-Diversity (2023)
Journal Article
Allard, M., Smith, S. C., Chatzilygeroudis, K., Lim, B., & Cully, A. (2023). Online Damage Recovery for Physical Robots with Hierarchical Quality-Diversity. ACM Transactions on Evolutionary Learning and Optimization, 3(2), Article 6. https://doi.org/10.1145/3596912

In real-world environments, robots need to be resilient to damages and robust to unforeseen scenarios. Quality-Diversity (QD) algorithms have been successfully used to make robots adapt to damages in seconds by leveraging a diverse set of learned ski... Read More about Online Damage Recovery for Physical Robots with Hierarchical Quality-Diversity.

Benchmarking Quality-Diversity Algorithms on Neuroevolution for Reinforcement Learning (2022)
Presentation / Conference Contribution
Flageat, M., Lim, B., Grillotti, L., Allard, M., Smith, S. C., & Cully, A. (2022, July). Benchmarking Quality-Diversity Algorithms on Neuroevolution for Reinforcement Learning. Paper presented at Gecco 2022, Boston, Massachusetts

We present a Quality-Diversity benchmark suite for Deep Neuroevolution in Reinforcement Learning domains for robot control. The suite includes the definition of tasks, environments, behavioral descriptors, and fitness. We specify different benchmarks... Read More about Benchmarking Quality-Diversity Algorithms on Neuroevolution for Reinforcement Learning.

Hierarchical quality-diversity for online damage recovery (2022)
Presentation / Conference Contribution
Allard, M., Smith, S. C., Chatzilygeroudis, K., & Cully, A. (2022, June). Hierarchical quality-diversity for online damage recovery. Presented at GECCO '22: Genetic and Evolutionary Computation Conference, Boston, Massachusetts

Adaptation capabilities, like damage recovery, are crucial for the deployment of robots in complex environments. Several works have demonstrated that using repertoires of pre-trained skills can enable robots to adapt to unforeseen mechanical damages... Read More about Hierarchical quality-diversity for online damage recovery.

Self-Explainable Robots in Remote Environments (2021)
Presentation / Conference Contribution
Chiyah Garcia, F. J., Smith, S. C., Lopes, J., Ramamoorthy, S., & Hastie, H. (2021, March). Self-Explainable Robots in Remote Environments. Presented at HRI '21: ACM/IEEE International Conference on Human-Robot Interaction, Boulder, Colorado

As robots and autonomous systems become more adept at handling complex scenarios, their underlying mechanisms also become increasingly complex and opaque. This lack of transparency can give rise to unverifiable behaviours, limiting the use of robots... Read More about Self-Explainable Robots in Remote Environments.

Counterfactual Explanation and Causal Inference In Service of Robustness in Robot Control (2020)
Presentation / Conference Contribution
Smith, S. C., & Ramamoorthy, S. (2020, October). Counterfactual Explanation and Causal Inference In Service of Robustness in Robot Control. Presented at 10th Joint IEEE International Conference on Development and Learning and Epigenetic Robotics 2020 - Virtual conference, Chile, Valparaiso, Chile

We propose an architecture for training generative models of counterfactual conditionals of the form, `can we modify event A to cause B instead of C?', motivated by applications in robot control. Using an `adversarial training' paradigm, an image-bas... Read More about Counterfactual Explanation and Causal Inference In Service of Robustness in Robot Control.

The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control (2020)
Journal Article
Smith, S. C., Dharmadi, R., Imrie, C., Si, B., & Herrmann, J. M. (2020). The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control. Frontiers in Neurorobotics, 14, Article 62. https://doi.org/10.3389/fnbot.2020.00062

The proposed architecture applies the principle of predictive coding and deep learning in a brain-inspired approach to robotic sensorimotor control. It is composed of many layers each of which is a recurrent network. The component networks can be spo... Read More about The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control.

Semi-supervised learning from demonstration through program synthesis: An inspection robot case study (2020)
Presentation / Conference Contribution
Smith, S. C., & Ramamoorthy, S. (2020, September). Semi-supervised learning from demonstration through program synthesis: An inspection robot case study. Presented at First Workshop on Agents and Robots for reliable Engineered Autonomy, Online

Semi-supervised learning improves the performance of supervised machine learning by leveraging methods from unsupervised learning to extract information not explicitly available in the labels. Through the design of a system that enables a robot to le... Read More about Semi-supervised learning from demonstration through program synthesis: An inspection robot case study.

Decoupled Sampling-Based Motion Planning for Multiple Autonomous Marine Vehicles (2018)
Presentation / Conference Contribution
Volpi, N. C., Smith, S. C., Pascoal, A. M., Simetti, E., Turetta, A., Alibani, M., & Polani, D. (2018, October). Decoupled Sampling-Based Motion Planning for Multiple Autonomous Marine Vehicles. Presented at OCEANS 2018 MTS/IEEE Charleston, Charleston, SC

There is increasing interest in the deployment and operation of multiple autonomous marine vehicles (AMVs) for a number of challenging scientific and commercial operational mission scenarios. Some of the missions, such as geotechnical surveying and 3... Read More about Decoupled Sampling-Based Motion Planning for Multiple Autonomous Marine Vehicles.

Expanding the Active Inference Landscape: More Intrinsic Motivations in the Perception-Action Loop (2018)
Journal Article
Biehl, M., Guckelsberger, C., Salge, C., Smith, S. C., & Polani, D. (2018). Expanding the Active Inference Landscape: More Intrinsic Motivations in the Perception-Action Loop. Frontiers in Neurorobotics, 12, https://doi.org/10.3389/fnbot.2018.00045

Active inference is an ambitious theory that treats perception, inference, and action selection of autonomous agents under the heading of a single principle. It suggests biologically plausible explanations for many cognitive phenomena, including cons... Read More about Expanding the Active Inference Landscape: More Intrinsic Motivations in the Perception-Action Loop.

Evaluation of Internal Models in Autonomous Learning (2018)
Journal Article
Smith, S. C., & Herrmann, J. M. (2019). Evaluation of Internal Models in Autonomous Learning. IEEE Transactions on Cognitive and Developmental Systems, 11(4), 463-472. https://doi.org/10.1109/tcds.2018.2865999

Internal models (IMs) can represent relations between sensors and actuators in natural and artificial agents. In autonomous robots, the adaptation of IMs and the adaptation of the behavior are interdependent processes which have been studied under pa... Read More about Evaluation of Internal Models in Autonomous Learning.

Homeokinetic Reinforcement Learning (2012)
Presentation / Conference Contribution
Smith, S. C., & Herrmann, J. M. (2011, September). Homeokinetic Reinforcement Learning. Presented at First IAPR TC3 Workshop, PSL 2011, Ulm, Germany

In order to find a control policy for an autonomous robot by reinforcement learning, the utility of a behaviour can be revealed locally through a modulation of the motor command by probing actions. For robots with many degrees of freedom, this type o... Read More about Homeokinetic Reinforcement Learning.

Clustering-Based Searching and Navigation in an Online News Source (2006)
Presentation / Conference Contribution
Smith, S. C., & Rodríguez, M. A. (2006, April). Clustering-Based Searching and Navigation in an Online News Source. Presented at 28th European Conference on IR Research, ECIR 2006, London, UK

The growing amount of online news posted on the WWW demands new algorithms that support topic detection, search, and navigation of news documents. This work presents an algorithm for topic detection that considers the temporal evolution of news and t... Read More about Clustering-Based Searching and Navigation in an Online News Source.