Learning Behavior Hierarchies via High-Dimensional Sensor Projection

@inproceedings{Levy2013LearningBH,
  title={Learning Behavior Hierarchies via High-Dimensional Sensor Projection},
  author={Simon D. Levy and Suraj Bajracharya and Ross W. Gayler},
  booktitle={AAAI Workshop: Learning Rich Representations from Low-Level Sensors},
  year={2013}
}
We propose a knowledge-representation architecture allowing a robot to learn arbitrarily complex, hierarchical / symbolic relationships between sensors and actuators. These relationships are encoded in high-dimensional, low-precision vectors that are very robust to noise. Low-dimensional (single-bit) sensor values are projected onto the highdimensional representation space using low-precision random weights, and the appropriate actions are then computed using elementwise vector multiplication… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.

References

Publications referenced by this paper.
Showing 1-7 of 7 references

Learning Computing With Robots: Python + Scribbler or Scribbler2

  • D. Kumar
  • Institute for Personal Robots in Education.
  • 2011
1 Excerpt

Compositionality and biologically plausible models

  • T. Stewart, C. Eliasmith
  • Werning, M.; Hinzen, W.; and Machery, E., eds…
  • 2009
1 Excerpt

Vehicles: Experiments in Synthetic Psychology

  • V. Braitenberg
  • MIT Press.
  • 1984
2 Excerpts

Similar Papers

Loading similar papers…