Zero-Shot Stitching in Reinforcement Learning using Relative Representations.
Antonio Pio RicciardiValentino MaiorcaLuca MoschellaRiccardo MarinEmanuele RodolàPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- state space
- data sets
- model free
- optimal policy
- robot control
- learning process
- machine learning
- multi agent reinforcement learning
- reinforcement learning algorithms
- object recognition
- learning algorithm
- dynamic programming
- object detection
- multiresolution
- markov decision processes
- multi agent
- action selection
- multiscale
- action space
- function approximators
- multiple representations
- autonomous learning
- genetic algorithm
- image stitching
- distributed representations