| bibtype |
C -
Conference Paper (international conference)
|
| ARLID |
0549007 |
| utime |
20250123094015.7 |
| mtime |
20211202235959.9 |
| SCOPUS |
85114446331 |
| WOS |
000853014000027 |
| DOI |
10.1109/ISSC52156.2021.9467863 |
| title
(primary) (eng) |
Hierarchical Bayesian Transfer Learning Between a Pair of Kalman Filters |
| specification |
| page_count |
5 s. |
| media_type |
E |
|
| serial |
| ARLID |
cav_un_epca*0549104 |
| ISBN |
978-1-6654-3429-4 |
| title
|
Proceedings of the 32nd Irish Signals and Systems Conference (ISSC) 2021 |
| publisher |
| place |
Piscataway |
| name |
IEEE |
| year |
2021 |
|
|
| keyword |
fully probabilistic design |
| keyword |
hierarchical models |
| keyword |
Bayesian transfer learning |
| keyword |
randomized design |
| keyword |
Kalman filters |
| author
(primary) |
| ARLID |
cav_un_auth*0370767 |
| name1 |
Papež |
| name2 |
Milan |
| 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*0370768 |
| name1 |
Quinn |
| name2 |
Anthony |
| institution |
UTIA-B |
| full_dept (cz) |
Adaptivní systémy |
| full_dept |
Department of Adaptive Systems |
| department (cz) |
AS |
| department |
AS |
| country |
IE |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| source |
|
| cas_special |
| project |
| project_id |
GA18-15970S |
| agency |
GA ČR |
| country |
CZ |
| ARLID |
cav_un_auth*0362986 |
|
| abstract
(eng) |
Transfer learning strategies are typically designed in a deterministic manner, without processing uncertainty in the knowledge transfer mechanism. They also require the dependence between the participating learning procedures-Bayesian filters in this work-to be explicitly modelled. This letter develops an approach which relaxes both of these restrictive assumptions. We frame the proposed Bayesian transfer learning technique as fully probabilistic design of an unknown hierarchical probability distribution conditioned on knowledge in the form of an external probability distribution. This yields a randomized design around a base density for transfer learning which has been reported in previous work by the authors. In the Kalman filtering context, this hierarchical relaxation-which induces a knowledge-driven mixture state predictor-significantly improves tracking performance when compared to conventional transfer learning methods. |
| action |
| ARLID |
cav_un_auth*0418083 |
| name |
Irish Signals and Systems Conference (ISSC 2021) /23./ |
| dates |
20210610 |
| mrcbC20-s |
20210611 |
| place |
Athlone |
| country |
IE |
|
| RIV |
BC |
| FORD0 |
10000 |
| FORD1 |
10100 |
| FORD2 |
10102 |
| reportyear |
2022 |
| num_of_auth |
2 |
| presentation_type |
PR |
| inst_support |
RVO:67985556 |
| permalink |
http://hdl.handle.net/11104/0325119 |
| mrcbC61 |
1 |
| cooperation |
| ARLID |
cav_un_auth*0418084 |
| name |
Department of Electronic and Electrical Engineering Trinity College Dublin, the University of Dublin |
| country |
IE |
|
| confidential |
S |
| article_num |
9467863 |
| arlyear |
2021 |
| mrcbU14 |
85114446331 SCOPUS |
| mrcbU24 |
PUBMED |
| mrcbU34 |
000853014000027 WOS |
| mrcbU63 |
cav_un_epca*0549104 Proceedings of the 32nd Irish Signals and Systems Conference (ISSC) 2021 IEEE 2021 Piscataway 978-1-6654-3429-4 |
|