Abstract
In this paper, we propose an analog value associative memory using Restricted Boltzmann Machine (AVAM). Research on treating knowledge is becoming more and more important such as in natural language processing and computer vision fields. Associative memory plays an important role to store knowledge. First, we obtain distributed representation of words with analog values using word2vec. Then the obtained distributed representation is learned in the proposed AVAM. In the evaluation experiments, we found simple but very important phenomenon in word2vec method: almost all of the values in the generated vectors are small values. By applying traditional normalization method for each word vector, the performance of the proposed AVAM is largely improved. Detailed experimental evaluations are carried out to show superior performance of the proposed AVAM.
| Original language | English |
|---|---|
| Pages (from-to) | 60-66 |
| Number of pages | 7 |
| Journal | Journal of Advanced Computational Intelligence and Intelligent Informatics |
| Volume | 23 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - 2019 Jan 1 |
Fingerprint
Keywords
- Associative memory
- Restricted Boltzmann Machine
- Semantic network
- Word2vec
ASJC Scopus subject areas
- Human-Computer Interaction
- Computer Vision and Pattern Recognition
- Artificial Intelligence
Cite this
Analog value associative memory using restricted Boltzmann machine. / Tsutsui, Yuichiro; Hagiwara, Masafumi.
In: Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol. 23, No. 1, 01.01.2019, p. 60-66.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Analog value associative memory using restricted Boltzmann machine
AU - Tsutsui, Yuichiro
AU - Hagiwara, Masafumi
PY - 2019/1/1
Y1 - 2019/1/1
N2 - In this paper, we propose an analog value associative memory using Restricted Boltzmann Machine (AVAM). Research on treating knowledge is becoming more and more important such as in natural language processing and computer vision fields. Associative memory plays an important role to store knowledge. First, we obtain distributed representation of words with analog values using word2vec. Then the obtained distributed representation is learned in the proposed AVAM. In the evaluation experiments, we found simple but very important phenomenon in word2vec method: almost all of the values in the generated vectors are small values. By applying traditional normalization method for each word vector, the performance of the proposed AVAM is largely improved. Detailed experimental evaluations are carried out to show superior performance of the proposed AVAM.
AB - In this paper, we propose an analog value associative memory using Restricted Boltzmann Machine (AVAM). Research on treating knowledge is becoming more and more important such as in natural language processing and computer vision fields. Associative memory plays an important role to store knowledge. First, we obtain distributed representation of words with analog values using word2vec. Then the obtained distributed representation is learned in the proposed AVAM. In the evaluation experiments, we found simple but very important phenomenon in word2vec method: almost all of the values in the generated vectors are small values. By applying traditional normalization method for each word vector, the performance of the proposed AVAM is largely improved. Detailed experimental evaluations are carried out to show superior performance of the proposed AVAM.
KW - Associative memory
KW - Restricted Boltzmann Machine
KW - Semantic network
KW - Word2vec
UR - http://www.scopus.com/inward/record.url?scp=85060870796&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85060870796&partnerID=8YFLogxK
U2 - 10.20965/jaciii.2019.p0060
DO - 10.20965/jaciii.2019.p0060
M3 - Article
AN - SCOPUS:85060870796
VL - 23
SP - 60
EP - 66
JO - Journal of Advanced Computational Intelligence and Intelligent Informatics
JF - Journal of Advanced Computational Intelligence and Intelligent Informatics
SN - 1343-0130
IS - 1
ER -