Marius-Constantin Dinu — accepted AI-conference papers

2023

Addressing Parameter Choice Issues in Unsupervised Domain Adaptation by Aggregation

Marius-Constantin Dinu, Markus Holzleitner, Maximilian Beck, Hoan Duc Nguyen, Andrea Huber, Hamid Eghbal-zadeh +4

ICLR 2023top-5%

We study the problem of choosing algorithm hyper-parameters in unsupervised domain adaptation, i.e., with labeled data in a source domain and unlabeled data in a target domain, drawn from a different input distribution. We follow the strategy to compute several models using different hyper-parameter…

Cited by 15 Source ↗PDF ↗code ↗Scholar ↗

2022

Align-RUDDER: Learning From Few Demonstrations by Reward Redistribution

Vihang Patil, Markus Hofmarcher, Marius-Constantin Dinu, Matthias Dorfer, Patrick M Blies, Johannes Brandstetter +2

ICML 2022oral

Reinforcement learning algorithms require many samples when solving complex hierarchical tasks with sparse and delayed rewards. For such complex tasks, the recently proposed RUDDER uses reward redistribution to leverage steps in the Q-function that are associated with accomplishing sub-tasks. Howeve…

Cited by 64 Source ↗PDF ↗code ↗Scholar ↗

2021

The balancing principle for parameter choice in distance-regularized domain adaptation

Werner Zellinger, Natalia Shepeleva, Marius-Constantin Dinu, Hamid Eghbal-zadeh, Hoan Duc Nguyen, Bernhard Nessler +2

NeurIPS 2021poster

We address the unsolved algorithm design problem of choosing a justified regularization parameter in unsupervised domain adaptation. This problem is intriguing as no labels are available in the target domain. Our approach starts with the observation that the widely-used method of minimizing the sour…

Cited by 10 Source ↗PDF ↗code ↗Scholar ↗