We study the problem of representation learning in goal-conditioned hierarchical reinforcement learning.
CONTINUOUS CONTROL HIERARCHICAL REINFORCEMENT LEARNING REPRESENTATION LEARNING
Combining parameter noise with traditional RL methods allows to combine the best of both worlds.
Recently, researchers have made significant progress combining the advances in deep learning for learning feature representations with reinforcement learning.
A platform for Applied Reinforcement Learning (Applied RL)
We adapt the ideas underlying the success of Deep Q-Learning to the continuous action domain.
In this paper, we propose an actor ensemble algorithm, named ACE, for continuous control with a deterministic policy in reinforcement learning.
The DeepMind Control Suite is a set of continuous control tasks with a standardised structure and interpretable rewards, intended to serve as performance benchmarks for reinforcement learning agents.
In this work, we propose to apply trust region optimization to deep reinforcement learning using a recently proposed Kronecker-factored approximation to the curvature.
This paper presents an actor-critic deep reinforcement learning agent with experience replay that is stable, sample efficient, and performs remarkably well on challenging environments, including the discrete 57-game Atari domain and several continuous control problems.
Planning has been very successful for control tasks with known environment dynamics.