Mapping State Space using Landmarks for Universal Goal Reaching
This addresses the challenge of efficient goal-reaching in reinforcement learning for agents in complex environments, representing an incremental improvement over existing methods.
The paper tackles the problem of learning Universal Value Function Approximators (UVFAs) in large MDPs with sparse rewards, where estimating value functions for long-range goals is difficult, and proposes a hierarchical method using landmarks to improve exploration and routing, resulting in early achievement of long-range goals and better performance than standard RL algorithms on challenging tasks.
An agent that has well understood the environment should be able to apply its skills for any given goals, leading to the fundamental problem of learning the Universal Value Function Approximator (UVFA). A UVFA learns to predict the cumulative rewards between all state-goal pairs. However, empirically, the value function for long-range goals is always hard to estimate and may consequently result in failed policy. This has presented challenges to the learning process and the capability of neural networks. We propose a method to address this issue in large MDPs with sparse rewards, in which exploration and routing across remote states are both extremely challenging. Our method explicitly models the environment in a hierarchical manner, with a high-level dynamic landmark-based map abstracting the visited state space, and a low-level value network to derive precise local decisions. We use farthest point sampling to select landmark states from past experience, which has improved exploration compared with simple uniform sampling. Experimentally we showed that our method enables the agent to reach long-range goals at the early training stage, and achieve better performance than standard RL algorithms for a number of challenging tasks.