Curriculum reinforcement learning via constrained optimal transport

Pascal Klink, Haoyi Yang, Carlo D'Eramo, Joni Pajarinen, Jan Peters

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

12 Sitaatiot (Scopus)
44 Lataukset (Pure)

Abstrakti

Curriculum reinforcement learning (CRL) allows solving complex tasks by generating a tailored sequence of learning tasks, starting from easy ones and subsequently increasing their difficulty. Although the potential of curricula in RL has been clearly shown in a variety of works, it is less clear how to generate them for a given learning environment, resulting in a variety of methods aiming to automate this task. In this work, we focus on the idea of framing curricula as interpolations between task distributions, which has previously been shown to be a viable approach to CRL. Identifying key issues of existing methods, we frame the generation of a curriculum as a constrained optimal transport problem between task distributions. Benchmarks show that this way of curriculum generation can improve upon existing CRL methods, yielding high performance in a variety of tasks with different characteristics.
AlkuperäiskieliEnglanti
OtsikkoProceedings of the 39th International Conference on Machine Learning
KustantajaJMLR
Sivumäärä18
TilaJulkaistu - 2022
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaInternational Conference on Machine Learning - Baltimore, Yhdysvallat
Kesto: 17 heinäk. 202223 heinäk. 2022
Konferenssinumero: 39

Julkaisusarja

NimiProceedings of Machine Learning Research
KustantajaPMLR
Vuosikerta162
ISSN (elektroninen)2640-3498

Conference

ConferenceInternational Conference on Machine Learning
LyhennettäICML
Maa/AlueYhdysvallat
KaupunkiBaltimore
Ajanjakso17/07/202223/07/2022

Sormenjälki

Sukella tutkimusaiheisiin 'Curriculum reinforcement learning via constrained optimal transport'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä