The simple simulator we used in our recent experiment on intrinsic motivation is now available online! Generate visuo-motor trajectories on the fly using the python scripts from this github repository and the visuo-motor dataset published in this Zenodo repository!
Very happy to give a talk tomorrow on implementations of prediction error dynamics at the WS on Prediction Error organised by Bruno Lara and Alejandra Ciria at the Humboldt-Universität in Berlin!
See the agenda here
Check out my new work submitted to Adaptive Behaviours (Sage Journal) on Intrinsic motivation and episodic memories for robot exploration on high-dimensional sensory spaces: https://arxiv.org/abs/2001.01982
The proposed framework adopts deep convolutional neural networks and shallow networks to generate curiosity-driven behaviours in a robot. Online learning of such models is performed. An episodic memory system is used to face catastrophic forgetting issues, typically experienced when performing online updates of artificial neural networks.
I am organising a workshop on “Predictive processes for motor and cognitive development in robots” as part of my Marie Sklodowska Curie project “Predictive Robots” at the BioRobotics Institute at Sant’Anna featuring invited talks from Prof. Bruno Lara (UAEM Mexico) and Dr. Alejandra Ciria (UAEM Mexico).
Evidences from cognitive sciences and neurosciences suggest that several perceptual phenomena and cognitive capabilities would rely on processes of anticipation of sensorimotor activity.
Different accounts on computational models for predictive processes have been proposed in the literature. This workshop provides an overview of established frameworks of predictive internal models – which are based on the idea that perception is a bottom-up flow of information sourced from sensory receptors – and of more recent proposals on predictive processing – where perception relies on deviations from top-down cortical predictions.
Moreover, this workshops aims at fostering discussions on the role that predictive processes may have in robotics, from supporting motor control and adaptive behaviours to enabling cognitive development.
Organised by Dr. Guido Schillaci as part of the EU-H2020 Marie Sklodowska Curie project “Predictive Robots”, this workshop will feature talks from two invited speakers, Prof. Bruno Lara (UAEM, Mexico, and Alexander von Humboldt fellow) and Dr. Alejandra Ciria (UAEM and Humboldt-Universität zu Berlin), and from members of the BioRobotics Institute of the Scuola Superiore Sant’Anna.
Aula 3, The BioRobotics Institute, Viale Rinaldo Piaggio 34, Pontedera (Italy)
Friday, 22nd November 2019, 11:00am
11:00 Welcome by Guido Schillaci
11:10 Internal models and developmental robotics (Bruno Lara)
11:40 Internal models and prediction error dynamics (Alejandra Ciria)
12:10 Predictive processes and the minimal self (Guido Schillaci)
12:30 Combining prediction and adaptation for robot control (Lorenzo Vannucci)
12:50 Brain-inspired algorithms for robot control (Egidio Falotico)
The Predictive Robots project is now associated to the German DFG Priority Programme “The Active Self”, an interdisciplinary network of more than 20 projects investigating the sensorimotor grounding of the human minimal self!
ALIFE 2019 Conference Proceedings are published in MIT Press. Check out our new paper on “Ego-Noise Predictions for Echolocation in Wheeled Robots”, authored by Antonio Pico, me, Verena Hafner and Bruno Lara: https://www.mitpressjournals.org/doi/abs/10.1162/isal_a_00222
Happy that our paper on “An Interdisciplinary Overview of Developmental Indices and Behavioral Measures of the Minimal Self“, written with Yasmin Kim Georgie and Verena Hafner, is one of the five nominees for the best paper award at IEEE ICDL-Epirob!
Pre-print available here http://arxiv.org/abs/1907.00709
The EU-H2020 MSCA-IF Predictive Robots project is also about spreading the adoption of a developmental paradigm and of predictive processes for different robotic applications. Greenhouses are complex systems that can be controlled, and whose states – and those of their crops – can be measured with a rich variety of sensors. From this point of view, they are not much different than robots. So, why not applying the ideas of developmental robotics on innovative greenhouses, as well?
Luis Miranda and I presented a work on “Adaptive Architectures towards portability of Greenhouse Models” at Greensys 2019, in Angers, France.
Adaptive models, trained in an incremental fashion, will help transferring knowledge between lab facilities and production greenhouses!
Pre-print of the paper available soon!
The Predictive Robots project has started! Read more about it here!
Delighted that my proposal for a EU-H2020 Marie Skłodowska Curie Individual Fellowship on “Predictive Robots” has been accepted! The project will be hosted by the BioRobotics Institute of the Sant’Anna School of Advanced Studies (Pisa, Italy). More research on developmental robotics, predictive models and on the artificial self coming soon!