Towards Open-Ended Learning of Action Sequences with Hierarchical Predictive Self-Organization

Open-ended learning is fundamental in autonomous robotics for the incremental acquisition of knowledge through experience. However, most of the proposed computational models for action recognition do not account for incremental learning, but rather learn a batch of training actions without adapting to new inputs presented after training sessions. Therefore, this is the need to provide robots with the ability to incrementally process a set of available perceptual cues and to adapt their behavioural responses over time. In this work, we propose a neural network architecture with multilayer-predictive processing for incrementally learning action sequences. Our architecture comprises a hierarchy of selforganizing networks that progressively learn the spatiotemporal structure of the input using Hebbian-like plasticity. Along the hierarchical flow with increasingly larger temporal receptive fields, feedback connections from higher-order networks carry predictions of lower-level neural activation patterns, whereas feedforward connections convey residual errors between the predictions and the lower-level activity. This mechanism is used to modulate the amount of learning necessary to adapt to the dynamic input distribution and develop robust action representations. We present a simplified hierarchical architecture with two layers and describe a number of planned experiments for classifying human actions in an open-ended learning scenario.

@InProceedings{PW16a, 
 	 author =  {Parisi, German I. and Wermter, Stefan},  
 	 title = {Towards Open-Ended Learning of Action Sequences with Hierarchical Predictive Self-Organization}, 
 	 booktitle = {IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), Workshop on Behaviours Adaptation, Interaction and Learning for Assistive Robotics},
 	 editors = {},
 	 number = {},
 	 volume = {},
 	 pages = {},
 	 year = {2016},
 	 month = {Aug},
 	 publisher = {},
 	 doi = {}, 
 }