Exploring Uncertainty and Movement in Categorical Perception Using Robots

@inproceedings{Powell2016ExploringUA,
  title={Exploring Uncertainty and Movement in Categorical Perception Using Robots},
  author={Nathaniel Powell and Josh C. Bongard},
  booktitle={PPSN},
  year={2016}
}
Cognitive agents are able to perform categorical perception through physical interaction (active categorical perception; ACP), or passively at a distance (distal categorical perception; DCP). It is possible that the former scaffolds the learning of the latter. However, it is unclear whether ACP indeed scaffolds DCP in humans and animals, nor how a robot could be trained to likewise learn DCP from ACP. Here we demonstrate a method for doing so which involves uncertainty: robots are trained to… 

References

SHOWING 1-10 OF 12 REFERENCES
Active Categorical Perception of Object Shapes in a Simulated Anthropomorphic Robotic Arm
TLDR
The results show that the agents solve the categorization task in an effective and robust way by self-selecting the required information through action and by integrating experienced sensory-motor states over time.
The Utility of Evolving Simulated Robot Morphology Increases with Task Complexity for Object Manipulation
TLDR
This work shows that for an object manipulation task in which a simulated robot must accomplish one, two, or three objectives simultaneously, subjugating more aspects of the robot's morphology to selective pressure allows for the evolution of better robots as the number of objectives increases.
Morphological computation: A basis for the analysis of morphology and control requirements
  • C. Paul
  • Biology
    Robotics Auton. Syst.
  • 2006
An Embodied Approach for Evolving Robust Visual Classifiers
TLDR
This work demonstrates a robot that achieves more robust categorization when it evolves to use proprioceptive sensors and is then trained to rely increasingly on vision, compared to a similar robot that is trained to categorize only with visual sensors.
The Dynamics of Active Categorical Perception in an Evolved Model Agent
TLDR
A model agent whose "nervous system" was evolved using a genetic algorithm to catch circular objects and to avoid diamond-shaped ones is studied to illustrate how the perspective and tools of dynamical systems theory can be applied to the analysis of situated, embodied agents capable of minimally cognitive behavior.
Towards a theoretical foundation for morphological computation with compliant bodies
TLDR
This work presents a model for morphological computation with compliant bodies, where a precise mathematical characterization of the potential computational contribution of a complex physical body is feasible and demonstrates that simple generic models of physical bodies, based on mass-spring systems, can be used to implement complex nonlinear operators.
Cognition from the bottom up: on biological inspiration, body morphology, and soft materials
Information Self-Structuring: Key Principle for Learning and Development
  • M. Lungarella, O. Sporns
  • Biology, Psychology
    Proceedings. The 4nd International Conference on Development and Learning, 2005.
  • 2005
TLDR
It is shown how information theoretical measures can be used to quantify statistical structure in sensory and motor channels of a robot capable of saliency-driven, attention-guided behavior and the potential importance of such measures for understanding sensorimotor coordination in organisms and for robot design.
Elephants don't play chess
  • R. Brooks
  • Computer Science
    Robotics Auton. Syst.
  • 1990
Query by committee
TLDR
It is suggested that asymptotically finite information gain may be an important characteristic of good query algorithms, in which a committee of students is trained on the same data set.
...
...