| bibtype |
C -
Conference Paper (international conference)
|
| ARLID |
0534000 |
| utime |
20240103224647.4 |
| mtime |
20201105235959.9 |
| SCOPUS |
85098853951 |
| DOI |
10.1109/SMC42975.2020.9283093 |
| title
(primary) (eng) |
Similarity-based transfer learning of decision policies |
| specification |
| page_count |
8 s. |
| media_type |
E |
|
| serial |
| ARLID |
cav_un_epca*0534238 |
| ISBN |
978-1-7281-8527-9 |
| ISSN |
1062-922X |
| title
|
Proceedings of the IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS 2020 |
| page_num |
37-44 |
| publisher |
| place |
Piscataway |
| name |
IEEE |
| year |
2020 |
|
|
| keyword |
probabilistic model |
| keyword |
fully probabilistic design |
| keyword |
transfer learning |
| keyword |
closed-loop behavior |
| keyword |
Bayesian estimation |
| keyword |
sequential decision making |
| author
(primary) |
| ARLID |
cav_un_auth*0398726 |
| name1 |
Zugarová |
| name2 |
Eliška |
| institution |
UTIA-B |
| full_dept (cz) |
Adaptivní systémy |
| full_dept (eng) |
Department of Adaptive Systems |
| department (cz) |
AS |
| department (eng) |
AS |
| country |
CZ |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0101092 |
| name1 |
Guy |
| name2 |
Tatiana Valentine |
| institution |
UTIA-B |
| full_dept (cz) |
Adaptivní systémy |
| full_dept |
Department of Adaptive Systems |
| department (cz) |
AS |
| department |
AS |
| full_dept |
Department of Adaptive Systems |
| share |
50 |
| garant |
A |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| source |
|
| cas_special |
| project |
| project_id |
LTC18075 |
| agency |
GA MŠk |
| country |
CZ |
| ARLID |
cav_un_auth*0372050 |
|
| abstract
(eng) |
We consider a problem of learning decision policy from past experience available. Using the Fully Probabilistic Design (FPD) formalism, we propose a new general approach for finding a stochastic policy from the past data. The proposedapproach assigns degree of similarity to all of the past closed-loop behaviors. The degree of similarity expresses how close the current decision making task is to a past task. Then it is used by Bayesian estimation to learn an approximate optimal policy, which comprises the best past experience. The approach learns decision policy directly from the data without interacting with any supervisor/expert or using any reinforcement signal. The past experience may consider a decision objective different than the current one. Moreover the past decision policy need not to be optimal with respect to the past objective. We demonstrate our approach on simulated examples and show that the learned policy achieves better performance than optimal FPD policy whenever a mismodeling is present. |
| action |
| ARLID |
cav_un_auth*0398727 |
| name |
IEEE International Conference on Systems, Man and Cybernetics 2020 |
| dates |
20201011 |
| mrcbC20-s |
20201014 |
| place |
Toronto |
| country |
CA |
|
| RIV |
BB |
| FORD0 |
10000 |
| FORD1 |
10100 |
| FORD2 |
10103 |
| reportyear |
2021 |
| num_of_auth |
2 |
| presentation_type |
PR |
| inst_support |
RVO:67985556 |
| permalink |
http://hdl.handle.net/11104/0312464 |
| confidential |
S |
| arlyear |
2020 |
| mrcbU14 |
85098853951 SCOPUS |
| mrcbU24 |
PUBMED |
| mrcbU34 |
WOS |
| mrcbU63 |
cav_un_epca*0534238 Proceedings of the IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS 2020 978-1-7281-8527-9 1062-922X 37 44 Piscataway IEEE 2020 |
|