Outcome-Guided Counterfactuals for Reinforcement Learning Agents from a Jointly Trained Generative Latent Space.
Eric YehPedro SequeiraJesse HostetlerMelinda T. GervasioPublished in: CoRR (2022)
Keyphrases
- latent space
- reinforcement learning agents
- generative model
- transfer learning
- gaussian process latent variable models
- latent variables
- dynamic environments
- reinforcement learning
- gaussian process
- low dimensional
- probabilistic model
- dimensionality reduction
- lower dimensional
- high dimensional
- manifold learning
- parameter space
- multi agent environments
- state abstraction
- learning tasks
- data sets
- feature space
- learning algorithm
- distance metric
- matrix factorization
- multi agent
- prior knowledge
- bayesian framework
- gaussian processes
- topic models
- labeled data
- training set
- machine learning algorithms
- high dimensional spaces
- autonomous agents
- text categorization
- unsupervised learning
- text mining
- semi supervised
- multi agent systems
- data mining