Algorithms for Learning Value-Aligned Policies Considering Admissibility Relaxation.
Andrés Holgado-SánchezJoaquín AriasHolger BillhardtSascha OssowskiPublished in: VALE (2023)
Keyphrases
- orders of magnitude
- learning process
- learning algorithm
- knowledge acquisition
- online learning
- learning models
- significant improvement
- data sets
- prior knowledge
- learning systems
- noise tolerant
- active learning
- supervised learning
- unsupervised learning
- computational complexity
- markov decision processes
- objective function
- inductive inference
- automatically learned