From Continual Learning to Causal Discovery in Robotics

Luca Castri, Sariah Mghames, Nicola Bellotto
Proceedings of The First AAAI Bridge Program on Continual Causality, PMLR 208:85-91, 2023.

Abstract

Reconstructing accurate causal models of dynamic systems from time-series of sensor data is a key problem in many real-world scenarios. In this paper, we present an overview based on our experience about practical challenges that the causal analysis encounters when applied to autonomous robots and how Continual Learning (CL) could help to overcome them. We propose a possible way to leverage the CL paradigm to make causal discovery feasible for robotics applications where the computational resources are limited, while at the same time exploiting the robot as an active agent that helps to increase the quality of the reconstructed causal models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v208-castri23a, title = {From Continual Learning to Causal Discovery in Robotics}, author = {Castri, Luca and Mghames, Sariah and Bellotto, Nicola}, booktitle = {Proceedings of The First AAAI Bridge Program on Continual Causality}, pages = {85--91}, year = {2023}, editor = {Mundt, Martin and Cooper, Keiland W. and Dhami, Devendra Singh and Ribeiro, Adéle and Smith, James Seale and Bellot, Alexis and Hayes, Tyler}, volume = {208}, series = {Proceedings of Machine Learning Research}, month = {07--08 Feb}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v208/castri23a/castri23a.pdf}, url = {https://proceedings.mlr.press/v208/castri23a.html}, abstract = {Reconstructing accurate causal models of dynamic systems from time-series of sensor data is a key problem in many real-world scenarios. In this paper, we present an overview based on our experience about practical challenges that the causal analysis encounters when applied to autonomous robots and how Continual Learning (CL) could help to overcome them. We propose a possible way to leverage the CL paradigm to make causal discovery feasible for robotics applications where the computational resources are limited, while at the same time exploiting the robot as an active agent that helps to increase the quality of the reconstructed causal models.} }
Endnote
%0 Conference Paper %T From Continual Learning to Causal Discovery in Robotics %A Luca Castri %A Sariah Mghames %A Nicola Bellotto %B Proceedings of The First AAAI Bridge Program on Continual Causality %C Proceedings of Machine Learning Research %D 2023 %E Martin Mundt %E Keiland W. Cooper %E Devendra Singh Dhami %E Adéle Ribeiro %E James Seale Smith %E Alexis Bellot %E Tyler Hayes %F pmlr-v208-castri23a %I PMLR %P 85--91 %U https://proceedings.mlr.press/v208/castri23a.html %V 208 %X Reconstructing accurate causal models of dynamic systems from time-series of sensor data is a key problem in many real-world scenarios. In this paper, we present an overview based on our experience about practical challenges that the causal analysis encounters when applied to autonomous robots and how Continual Learning (CL) could help to overcome them. We propose a possible way to leverage the CL paradigm to make causal discovery feasible for robotics applications where the computational resources are limited, while at the same time exploiting the robot as an active agent that helps to increase the quality of the reconstructed causal models.
APA
Castri, L., Mghames, S. & Bellotto, N.. (2023). From Continual Learning to Causal Discovery in Robotics. Proceedings of The First AAAI Bridge Program on Continual Causality, in Proceedings of Machine Learning Research 208:85-91 Available from https://proceedings.mlr.press/v208/castri23a.html.

Related Material