A loosely collaborative dependency framework for a fast adaptive agent model using extended RSVD
This paper proposes a loosely dependent agent model, where the underlying training is based on a collaborative filtering, i.e., an extended Regularized Singular Value Decomposition (RSVD) technique. Such software agents normally act as a cooperative assistant to carry out some tasks on behalf of the...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
Indian Society for Development and Environment Research
2015
|
| Subjects: | |
| Online Access: | http://www.scopus.com/inward/record.url?partnerID=HzOxMe3b&scp=84879751476&origin=inward http://cmuir.cmu.ac.th/handle/6653943832/38694 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Chiang Mai University |
| Summary: | This paper proposes a loosely dependent agent model, where the underlying training is based on a collaborative filtering, i.e., an extended Regularized Singular Value Decomposition (RSVD) technique. Such software agents normally act as a cooperative assistant to carry out some tasks on behalf of their user, thereby operate alone smoothly in any virtual environment. The extended RSVD technique runs periodically at the server site to rapidly update the agent's experiences. The derived knowledge in conjunction with those from other collaborative agents are encapsulated as prior knowledge and sent back to the client agent. Thus, the agent can execute its tasks having timely repertoire of experience to make suitable decisions. The adaptation is said to complete. © 2013 by IJAI. |
|---|