0000000000768244
AUTHOR
Miriam Buonamente
Simulating music with associative self-organizing maps
Abstract We present an architecture able to recognise pitches and to internally simulate likely continuations of partially heard melodies. Our architecture consists of a novel version of the Associative Self-Organizing Map (A-SOM) with generalized ancillary connections. We tested the performance of our architecture with melodies from a publicly available database containing 370 Bach chorale melodies. The results showed that the architecture could learn to represent and perfectly simulate the remaining 20% of three different interrupted melodies when using a context length of 8 centres of activity in the A-SOM. These promising and encouraging results show that our architecture offers somethi…
Recognizing actions with the associative self-organizing map
When artificial agents interact and cooperate with other agents, either human or artificial, they need to recognize others’ actions and infer their hidden intentions from the sole observation of their surface level movements. Indeed, action and intention understanding in humans is believed to facilitate a number of social interactions and is supported by a complex neural substrate (i.e. the mirror neuron system). Implementation of such mechanisms in artificial agents would pave the route to the development of a vast range of advanced cognitive abilities, such as social interaction, adaptation, and learning by imitation, just to name a few. We present a first step towards a fully-fledged int…
Hierarchies of Self-Organizing Maps for action recognition
We propose a hierarchical neural architecture able to recognise observed human actions. Each layer in the architecture represents increasingly complex human activity features. The first layer consists of a SOM which performs dimensionality reduction and clustering of the feature space. It represents the dynamics of the stream of posture frames in action sequences as activity trajectories over time. The second layer in the hierarchy consists of another SOM which clusters the activity trajectories of the first-layer SOM and learns to represent action prototypes. The third - and last - layer of the hierarchy consists of a neural network that learns to label action prototypes of the second-laye…
Simulating Actions with the Associative Self-Organizing Map
We present a system that can learn to represent actions as well as to internally simulate the likely continuation of their initial parts. The method we propose is based on the Associative Self Organizing Map (A-SOM), a variant of the Self Organizing Map. By emulating the way the human brain is thought to perform pattern recognition tasks, the A- SOM learns to associate its activity with di erent inputs over time, where inputs are observations of other's actions. Once the A-SOM has learnt to recognize actions, it uses this learning to predict the continuation of an observed initial movement of an agent, in this way reading its intentions. We evaluate the system's ability to simulate actions …
Internal Simulation of an Agent’s Intentions
We present the Associative Self-Organizing Map (A-SOM) and propose that it could be used to predict an agent’s intentions by internally simulating the behaviour likely to follow initial movements. The A-SOM is a neural network that develops a representation of its input space without supervision, while simultaneously learning to associate its activity with an arbitrary number of additional (possibly delayed) inputs. We argue that the A-SOM would be suitable for the prediction of the likely continuation of the perceived behaviour of an agent by learning to associate activity patterns over time, and thus a way to read its intentions.
Action Recognition based on Hierarchical Self-Organizing Maps
We propose a hierarchical neural architecture able to recognise observed human actions. Each layer in the architecture represents increasingly complex human activity features. The first layer consists of a SOM which performs dimensionality reduction and clustering of the feature space. It represents the dynamics of the stream of posture frames in action sequences as activity trajectories over time. The second layer in the hierarchy consists of another SOM which clusters the activity trajectories of the first-layer SOM and thus it learns to represent action prototypes independent of how long the activity trajectories last. The third layer of the hierarchy consists of a neural network that le…
Architecture to Serve Disabled and Elderly
We propose an architecture (discussed in the context of a dressing and cleaning application for impaired and elderly persons) that combines a cognitive framework that generates motor commands with the MOSAIC architecture which selects the right motor command according to the proper context. The ambition is to have robots able to understand humans intentions (dressing or cleaning intentions), to learn new tasks only by observing humans, and to represent the world around it by using conceptual spaces. The cognitive framework implements the learning by demonstration paradigm and solves the related problem to map the observed movement into the robot motor system. Such framework is assumed to wo…
Discriminating and simulating actions with the associative self-organising map
We propose a system able to represent others’ actions as well as to internally simulate their likely continuation from a partial observation. The approach presented here is the first step towards a more ambitious goal of endowing an artificial agent with the ability to recognise and predict others’ intentions. Our approach is based on the associative self-organising map, a variant of the self-organising map capable of learning to associate its activity with different inputs over time, where inputs are processed observations of others’ actions. We have evaluated our system in two different experimental scenarios obtaining promising results: the system demonstrated an ability to learn discrim…