Policy transfer via modularity and reward guiding

41 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Driving Policy Transfer via Modularity and Abstraction (2018, background): “irect sim-to-real generalization by maximizing the diversity of textures and occluders in simulation. Other approaches tackle transfer via modularization in the context of manual grasping and pushing [10,7]. Clavera et al. [7] propose an approach that is conceptually similar to ours in that a system”
Modular Networks: Learning to Decompose Neural Computation (2018, background): “There is also a wide range of literature in robotics that uses modularity to learn robot skills more efficiently by reusing functionality shared between tasks [19, 6].”
DREAM Architecture: a Developmental Approach to Open-Ended Learning in Robotics (2020, unknown): “Transfer learning is also particularly interesting to learn in simulation before transferring to reality, as it drastically reduces the number of required samples on the real robot [16, 17, 18, 19, 20].”
Transfer in Deep Reinforcement Learning Using Successor Features and Generalised Policy Improvement (2018, background): “Many works on the combination of deep RL and transfer propose modular network architectures that naturally induce a decomposition of the problem (Devin et al., 2016; Heess et al., 2017; Clavera et al., 2017).”