Progressive growing of self-organized hierarchical representations for exploration

Abstract

Designing agent that can autonomously discover and learn a diversity of structures and skills in unknown changing environments is key for lifelong machine learning. A central challenge is how to learn incrementally representations in order to progressively build a map of the discovered structures and re-use it to further explore. To address this challenge, we identify and target several key functionalities. First, we aim to build lasting representations and avoid catastrophic forgetting throughout the exploration process. Secondly we aim to learn a diversity of representations allowing to discover a "diversity of diversity" of structures (and associated skills) in complex high-dimensional environments. Thirdly, we target representations that can structure the agent discoveries in a coarse-to-fine manner. Finally, we target the reuse of such representations to drive exploration toward an "interesting" type of diversity, for instance leveraging human guidance. Current approaches in state representation learning rely generally on monolithic architectures which do not enable all these functionalities. Therefore, we present a novel technique to progressively construct a Hierarchy of Observation Latent Models for Exploration Stratification, called HOLMES. This technique couples the use of a dynamic modular model architecture for representation learning with intrinsically-motivated goal exploration processes (IMGEPs). The paper shows results in the domain of automated discovery of diverse self-organized patterns, considering as testbed the experimental framework from Reinke et al. (2019).

References in corpus (9)

• How transferable are features in deep neural networks?
• Recurrent World Models Facilitate Policy Evolution
• Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning
• Intrinsically Motivated Goal Exploration Processes with Automatic Curriculum Learning
• Learning Plannable Representations with Causal InfoGAN
• Skew-Fit: State-Covering Self-Supervised Reinforcement Learning
• Unsupervised Learning of Goal Spaces for Intrinsically Motivated Goal Exploration
• Continual State Representation Learning for Reinforcement Learning using Generative Replay
• Contextual Imagined Goals for Self-Supervised Robotic Learning