Sensorimotor experience and its metrics: informational geometry and the temporal horizon

C.L. Nehaniv

Sensorimotor experience and its metrics: informational geometry and the temporal horizon

C.L. Nehaniv

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

7 Citations (Scopus)

23 Downloads (Pure)

Abstract

We introduce metrics on sensorimotor experience
at various temporal scales based on informationtheory.
Sensorimotor variables through which the experience
of an agent flows are modeled as information
sources in the sense of Shannon information theory. Information
distance between the constellation of an embodied
agent's sensorimotor variables at different moments
in time can be taken variable-by-variable or between
entire sets of such variables to yield two classes of
metrics on sensorimotor experience: the temporal experiential
information distance and the Hausdorff metric on
experience. Unlike mutual information, these measures
each satisfy the metric axioms and thus induce a geometry
on the space of experiences with the same temporal
scope. Continuity of maps between experiential spaces
as well as robotic applications and extensions are discussed.

Original language	English
Title of host publication	Procs of 2005 IEEE Congress on Evolutionary Computation
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	142-149
Volume	1
ISBN (Print)	0-7803-9363-5
Publication status	Published - 2005

Access to Document

901934
“This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder." “Copyright IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.”
Final published version, 2.73 MB

Cite this

@inproceedings{3ec66e7dcf8c4d8fa3b5806d29d31748,

title = "Sensorimotor experience and its metrics: informational geometry and the temporal horizon",

abstract = "We introduce metrics on sensorimotor experience at various temporal scales based on informationtheory. Sensorimotor variables through which the experience of an agent flows are modeled as information sources in the sense of Shannon information theory. Information distance between the constellation of an embodied agent's sensorimotor variables at different moments in time can be taken variable-by-variable or between entire sets of such variables to yield two classes of metrics on sensorimotor experience: the temporal experiential information distance and the Hausdorff metric on experience. Unlike mutual information, these measures each satisfy the metric axioms and thus induce a geometry on the space of experiences with the same temporal scope. Continuity of maps between experiential spaces as well as robotic applications and extensions are discussed.",

author = "C.L. Nehaniv",

note = "“This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.{"} “Copyright IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.”",

year = "2005",

language = "English",

isbn = "0-7803-9363-5",

volume = "1",

pages = "142--149",

booktitle = "Procs of 2005 IEEE Congress on Evolutionary Computation",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

}

TY - GEN

T1 - Sensorimotor experience and its metrics: informational geometry and the temporal horizon

AU - Nehaniv, C.L.

N1 - “This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder." “Copyright IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.”

PY - 2005

Y1 - 2005

N2 - We introduce metrics on sensorimotor experience at various temporal scales based on informationtheory. Sensorimotor variables through which the experience of an agent flows are modeled as information sources in the sense of Shannon information theory. Information distance between the constellation of an embodied agent's sensorimotor variables at different moments in time can be taken variable-by-variable or between entire sets of such variables to yield two classes of metrics on sensorimotor experience: the temporal experiential information distance and the Hausdorff metric on experience. Unlike mutual information, these measures each satisfy the metric axioms and thus induce a geometry on the space of experiences with the same temporal scope. Continuity of maps between experiential spaces as well as robotic applications and extensions are discussed.

AB - We introduce metrics on sensorimotor experience at various temporal scales based on informationtheory. Sensorimotor variables through which the experience of an agent flows are modeled as information sources in the sense of Shannon information theory. Information distance between the constellation of an embodied agent's sensorimotor variables at different moments in time can be taken variable-by-variable or between entire sets of such variables to yield two classes of metrics on sensorimotor experience: the temporal experiential information distance and the Hausdorff metric on experience. Unlike mutual information, these measures each satisfy the metric axioms and thus induce a geometry on the space of experiences with the same temporal scope. Continuity of maps between experiential spaces as well as robotic applications and extensions are discussed.

M3 - Conference contribution

SN - 0-7803-9363-5

VL - 1

SP - 142

EP - 149

BT - Procs of 2005 IEEE Congress on Evolutionary Computation

PB - Institute of Electrical and Electronics Engineers (IEEE)

ER -

Sensorimotor experience and its metrics: informational geometry and the temporal horizon

Abstract

Access to Document

Fingerprint

Cite this