Tracking emotions: intrinsic motivation grounded on multi-level prediction error dynamics

Happy that my paper on “Tracking emotions: intrinsic motivation grounded on multi-level prediction error dynamics”, co-authored with Alejandra Ciria (UNAM, MX) and Bruno Lara (UAEM, MX), has been accepted for presentation at IEEE ICDL-Epirob 2020!

In this work, we propose a learning architecture that generates exploratory behaviours towards self-generated goals in a simulated robot, and that regulates goal selection and the balance between exploitation and exploration through a multi-level monitoring of prediction error dynamics.

The system is made of: 1) a convolutional autoencoder for unsupervised learning of low-dimensional features from visual inputs; 2) a self-organising map for online learning of visual goals; two deep neural networks, trained in an online fashion, encoding controller and predictor of the system. Memory replay is employed to face catastrophic forgetting issues.

A multi-level monitoring mechanism keeps track of two errors: (1) a high-level, general error of the system, i.e. MSE of the forward model calculated on a test dataset; (2) low-level goal errors, i.e. the prediction errors estimated when trying to reach each specific goal.

The system maintains a buffer of high-level MSE observed during a specific time window. After every update of the MSE buffer, a linear regression is calculated on the stored values over time, whose slope indicates the trend of the general error of the system.

This trend modulates computational resources (size of goal error buffers) and exploration noise: when overall performances improve, the necessity of tracking the goal error dynamics is reduced. On the contrary, the system widens the time window on which goal errors are monitored.

We discuss the tight relationship that PE dynamics may have with the emotional valence of action. PE dynamics may be fundamental cause of emotional valence of action: positive valence linked to an active reduction of PE and a negative valence to a continuous increase of PE.

Read the full paper here: https://arxiv.org/abs/2007.14632 (pre-print)!

Prediction-error driven memory consolidation for continual learning and adaptive greenhouse models

Check my AI Transfer work submitted to Springer KI (German Journal on Artificial Intelligence, special issue on Developmental Robotics) on “Prediction error-driven memory consolidation for continual learning”, applied on data from innovative greenhouses: https://arxiv.org/abs/2006.12616.

Episodic memory replay and prediction-error driven consolidation are used to tackle online learning in deep recurrent neural networks. Inspired by evidences from cognitive sciences and neuroscience, memories are retained depending on their congruency with prior knowledge.
This congruency is estimated in terms of prediction errors resulting from a generative model. In particular, our framework chooses which samples to maintain in the episodic memory based on their expected contribution to the learning progress.
Different retention strategies are compared. We analyse their impact on the variance of the samples stored in the memory and on the stability/plasticity of the model.

Co-authored with Luis Miranda and Uwe Schmidt, Humboldt-Universität zu Berlin.

SAGE Adaptive Behavior

Our article on “Intrinsic Motivation and Episodic Memories for Robot Exploration of High-Dimensional Sensory Spaces” is out in SAGE Adaptive Behaviour! Pre-print available here: https://arxiv.org/abs/2001.01982

Co-authored with Antonio Pico (HU-Berlin), Verena Hafner (HU-Berlin), Peter Hanappe (Sony CSL), David Colliaux (Sony CSL) and Timothee Wintz (Sony CSL)

Intrinsic motivation and episodic memories for robot exploration on high dimensional sensory spaces

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.

Workshop on “Predictive processes for motor and cognitive development in robots” at Sant’Anna

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).

Download the flyer here. Download the full programme here.

Details follows!

Abstract

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.

Where: 

Aula 3, The BioRobotics Institute, Viale Rinaldo Piaggio 34, Pontedera (Italy)

When:

Friday, 22nd November 2019, 11:00am

Agenda

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)

13:10 Conclusions