Influence of Clustal W hyperparameters in multiple sequences alignment for AlignRUDDER

Marlene GANZ

Bakalářská práce

ClustalW, Multiple Sequence Alignment, Reinforcement learning, Influence of Clustal W hyperparameters in mutiple sequences alignment for AlignRUDDER

Abstract:

Reinforcement learning algorithms suffer from the delayed reward problem and usually only perform well when being trained with vast amounts of data. AlignRUDDER overcomes this problem by using a multiple sequence alignment for the initialization performed with ClustalW. In this thesis we work with different data­sets and AlignRUDDER to search for optimal align­ ment hyperparameters for reinforcement learning problems.

Abstract:

Reinforcement learning algorithms suffer from the delayed reward problem and usually only perform well when being trained with vast amounts of data. AlignRUDDER overcomes this problem by using a multiple sequence alignment for the initialization performed with ClustalW. In this thesis we work with different data­sets and AlignRUDDER to search for optimal align­ ment hyperparameters for reinforcement learning problems.

Keywords

ClustalW, Reinforcement Learning, AlignRUDDER, Multiple Sequence Alignment, Hyperparameter Search

Jazyk práce: angličtina

Datum vytvoření / odevzdání či podání práce: 17. 3. 2021

Obhajoba závěrečné práce

• Vedoucí: prof. Dr. Sepp Hochreiter

JIHOČESKÁ UNIVERZITA V ČESKÝCH BUDĚJOVICÍCH

Přírodovědecká fakulta

Bakalářský studijní program / obor:
Applied Informatics / Bioinformatics