| bibtype |
A -
Abstract
|
| ARLID |
0616996 |
| utime |
20250224135810.4 |
| mtime |
20250217235959.9 |
| title
(primary) (eng) |
Exploration in Reinforcement Learning |
| specification |
| page_count |
1 s. |
| media_type |
P |
|
| serial |
| ARLID |
cav_un_epca*0604531 |
| title
|
DYNALIFE WG1-WG2 Interaction Meeting Data driven evidence: theoretical models and complex biological data |
| page_num |
12-12 |
| publisher |
| place |
Brusel |
| name |
The European Cooperation in Science and Technology (COST) |
| year |
2024 |
|
|
| keyword |
Reinforcement Learning |
| keyword |
Markov Decision Processes |
| keyword |
Exploration methods |
| author
(primary) |
| ARLID |
cav_un_auth*0483444 |
| name1 |
Jedlička |
| name2 |
Adam |
| institution |
UTIA-B |
| full_dept (cz) |
Adaptivní systémy |
| full_dept (eng) |
Department of Adaptive Systems |
| department (cz) |
AS |
| department (eng) |
AS |
| country |
CZ |
| share |
100 |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| source |
|
| source |
|
| cas_special |
| project |
| project_id |
CA21169 |
| agency |
EU-COST |
| country |
XE |
| ARLID |
cav_un_auth*0452289 |
|
| abstract
(eng) |
The so-called exploration-exploitation dilemma refers to optimizing the trade-off between discovering new states (exploration) and using already gathered knowledge for immediate reward The importance of the proper choice of this exploration algorithm lies in the potentially large improvement in the speed of convergence of the RL algorithm. The choice of a well-performing exploration algorithm is task and domain-specific thus there is no universal algorithm that would perform the best for every given task. The proposed poster will i) briefly introduce a mechanism of how RL works along with the comprehensive implementation of a biology-related task into an MDP that is suitable to be solved by RL. ii) describe several exploration algorithms (from rather simple ε-greedy exploration to more complex methods such as the Intrinsic Curiosity Module (ICM)) along with their benefits and show how exactly they fit into the overall RL mechanis. |
| action |
| ARLID |
cav_un_auth*0481157 |
| name |
DYNALIFE Interaction Meeting Data driven evidence: theoretical models and complex biological data |
| dates |
20240605 |
| country |
GR |
| mrcbC20-s |
20240607 |
| place |
Thessaloniki |
|
| RIV |
IN |
| FORD0 |
10000 |
| FORD1 |
10200 |
| FORD2 |
10201 |
| reportyear |
2025 |
| num_of_auth |
1 |
| presentation_type |
PO |
| inst_support |
RVO:67985556 |
| permalink |
https://hdl.handle.net/11104/0364271 |
| confidential |
S |
| arlyear |
2024 |
| mrcbU02 |
A2 |
| mrcbU14 |
SCOPUS |
| mrcbU24 |
PUBMED |
| mrcbU34 |
WOS |
| mrcbU63 |
cav_un_epca*0604531 DYNALIFE WG1-WG2 Interaction Meeting Data driven evidence: theoretical models and complex biological data 12 12 Brusel The European Cooperation in Science and Technology (COST) 2024 |
|