Lupa

Izpis gradiva Pomoč

A- | A+ | Natisni
Naslov:Qualitative control learning can be much faster than reinforcement learning
Avtorji:ID Šoberl, Domen (Avtor)
ID Bratko, Ivan (Avtor)
Datoteke:.pdf RAZ_Soberl_Domen_2025.pdf (1,53 MB)
MD5: 0481ED74AFC01EA22EAC32A81FB7E6DA
 
URL https://link.springer.com/article/10.1007/s10994-024-06724-7
 
Jezik:Angleški jezik
Vrsta gradiva:Članek v reviji
Tipologija:1.01 - Izvirni znanstveni članek
Organizacija:FAMNIT - Fakulteta za matematiko, naravoslovje in informacijske tehnologije
Opis:Reinforcement learning has emerged as a prominent method for controlling dynamic systems in the absence of a precise mathematical model. However, its reliance on extensive interactions with the environment often leads to prolonged training periods. In this paper, we propose an alternative approach to learning control policies that focuses on learning qualitative models and uses symbolic planning to derive a qualitative plan for the control task, which is executed by an adaptive reactive controller. We conduct experiments utilizing our approach on the cart-pole problem, a standard benchmark in dynamic system control. We additionally extend this problem domain to include uneven terrains, such as driving over craters or hills, to assess the robustness of learned controllers. Our results indicate that qualitative learning offers significant advantages over reinforcement learning in terms of sample efficiency, transferability, and interpretability. We demonstrate that our proposed approach is at least two orders of magnitude more sample efficient in the cart-pole domain than the usual variants of reinforcement learning.
Ključne besede:qualitative modeling, qualitative reasoning, qualitative control, transfer learning
Datum objave:14.01.2025
Leto izida:2025
Št. strani:str. 1-21
Številčenje:Vol. 114, article no. ǂ4
PID:20.500.12556/RUP-21533 Povezava se odpre v novem oknu
UDK:004.85
ISSN pri članku:0885-6125
DOI:10.1007/s10994-024-06724-7 Povezava se odpre v novem oknu
COBISS.SI-ID:222246659 Povezava se odpre v novem oknu
Datum objave v RUP:07.08.2025
Število ogledov:540
Število prenosov:8
Metapodatki:XML DC-XML DC-RDF
:
Kopiraj citat
  
Skupna ocena:(0 glasov)
Vaša ocena:Ocenjevanje je dovoljeno samo prijavljenim uporabnikom.
Objavi na:Bookmark and Share


Postavite miškin kazalec na naslov za izpis povzetka. Klik na naslov izpiše podrobnosti ali sproži prenos.

Gradivo je del revije

Naslov:Machine learning
Skrajšan naslov:Mach. learn.
Založnik:Kluwer Academic Publishers
ISSN:0885-6125
COBISS.SI-ID:2623527 Povezava se odpre v novem oknu

Gradivo je financirano iz projekta

Financer:ARIS - Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije
Številka projekta:P2-0209-2022
Naslov:Umetna inteligenca in inteligentni sistemi

Sekundarni jezik

Jezik:Slovenski jezik
Ključne besede:kvalitativno modeliranje, kvalitativno sklepanje, kvalitativno vodenje, prenosno učenje


Komentarji

Dodaj komentar

Za komentiranje se morate prijaviti.

Komentarji (0)
0 - 0 / 0
 
Ni komentarjev!

Nazaj
Logotipi partnerjev Univerza v Mariboru Univerza v Ljubljani Univerza na Primorskem Univerza v Novi Gorici