Unsupervised coding with lococode

Sepp Hochreiter; Jürgen Schmidhuber

doi:10.1007/bfb0020229

Unsupervised coding with lococode

Sepp Hochreiter, Jürgen Schmidhuber

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

5 Scopus citations

Abstract

Traditional approaches to sensory coding use code component-oriented objective functions (COCOFs) to evaluate code quality. Previous COCOFs do not take into account the information-theoretic complexity of the code-generating mapping itself. We do: “Low-complexity coding and decoding” (Lococode) generates so-called lococodes that (1) convey information about the input data, (2) can be computed from the data by a low-complexity mapping (LCM), and (3) can be decoded by a lcm. We implement lococode by training autoassociators with Flat Minimum Search (FMS), a general method for finding low-complexity neural nets. lococode extracts optimal codes for difficult versions of the "bars" benchmark problem. As a preprocessor for a vowel recognition benchmark problem it sets the stage for excellent classification performance.

Original language	English (US)
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher	Springer Verlag
Pages	655-660
Number of pages	6
ISBN (Print)	3540636315
DOIs	https://doi.org/10.1007/bfb0020229
State	Published - Jan 1 1997
Externally published	Yes

Access to Document

10.1007/bfb0020229

Cite this

@inproceedings{7c20185fed0a49b29795a3110550a7e2,

title = "Unsupervised coding with lococode",

abstract = "Traditional approaches to sensory coding use code component-oriented objective functions (COCOFs) to evaluate code quality. Previous COCOFs do not take into account the information-theoretic complexity of the code-generating mapping itself. We do: “Low-complexity coding and decoding” (Lococode) generates so-called lococodes that (1) convey information about the input data, (2) can be computed from the data by a low-complexity mapping (LCM), and (3) can be decoded by a lcm. We implement lococode by training autoassociators with Flat Minimum Search (FMS), a general method for finding low-complexity neural nets. lococode extracts optimal codes for difficult versions of the {"}bars{"} benchmark problem. As a preprocessor for a vowel recognition benchmark problem it sets the stage for excellent classification performance.",

author = "Sepp Hochreiter and J{\"u}rgen Schmidhuber",

note = "Generated from Scopus record by KAUST IRTS on 2022-09-14",

year = "1997",

month = jan,

day = "1",

doi = "10.1007/bfb0020229",

language = "English (US)",

isbn = "3540636315",

pages = "655--660",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

address = "Germany",

}

TY - GEN

T1 - Unsupervised coding with lococode

AU - Hochreiter, Sepp

AU - Schmidhuber, Jürgen

N1 - Generated from Scopus record by KAUST IRTS on 2022-09-14

PY - 1997/1/1

Y1 - 1997/1/1

N2 - Traditional approaches to sensory coding use code component-oriented objective functions (COCOFs) to evaluate code quality. Previous COCOFs do not take into account the information-theoretic complexity of the code-generating mapping itself. We do: “Low-complexity coding and decoding” (Lococode) generates so-called lococodes that (1) convey information about the input data, (2) can be computed from the data by a low-complexity mapping (LCM), and (3) can be decoded by a lcm. We implement lococode by training autoassociators with Flat Minimum Search (FMS), a general method for finding low-complexity neural nets. lococode extracts optimal codes for difficult versions of the "bars" benchmark problem. As a preprocessor for a vowel recognition benchmark problem it sets the stage for excellent classification performance.

AB - Traditional approaches to sensory coding use code component-oriented objective functions (COCOFs) to evaluate code quality. Previous COCOFs do not take into account the information-theoretic complexity of the code-generating mapping itself. We do: “Low-complexity coding and decoding” (Lococode) generates so-called lococodes that (1) convey information about the input data, (2) can be computed from the data by a low-complexity mapping (LCM), and (3) can be decoded by a lcm. We implement lococode by training autoassociators with Flat Minimum Search (FMS), a general method for finding low-complexity neural nets. lococode extracts optimal codes for difficult versions of the "bars" benchmark problem. As a preprocessor for a vowel recognition benchmark problem it sets the stage for excellent classification performance.

UR - http://link.springer.com/10.1007/BFb0020229

UR - http://www.scopus.com/inward/record.url?scp=84956696675&partnerID=8YFLogxK

U2 - 10.1007/bfb0020229

DO - 10.1007/bfb0020229

M3 - Conference contribution

SN - 3540636315

SP - 655

EP - 660

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -

Unsupervised coding with lococode

Abstract

Access to Document

Other files and links

Fingerprint

Cite this