Sub-Goal Trees -- a Framework for Goal-Based Reinforcement Learning
About
Many AI problems, in robotics and other domains, are goal-based, essentially seeking trajectories leading to various goal states. Reinforcement learning (RL), building on Bellman's optimality equation, naturally optimizes for a single goal, yet can be made multi-goal by augmenting the state with the goal. Instead, we propose a new RL framework, derived from a dynamic programming equation for the all pairs shortest path (APSP) problem, which naturally solves multi-goal queries. We show that this approach has computational benefits for both standard and approximate dynamic programming. Interestingly, our formulation prescribes a novel protocol for computing a trajectory: instead of predicting the next state given its predecessor, as in standard RL, a goal-conditioned trajectory is constructed by first predicting an intermediate state between start and goal, partitioning the trajectory into two. Then, recursively, predicting intermediate points on each sub-segment, until a complete trajectory is obtained. We call this trajectory structure a sub-goal tree. Building on it, we additionally extend the policy gradient methodology to recursively predict sub-goals, resulting in novel goal-based algorithms. Finally, we apply our method to neural motion planning, where we demonstrate significant improvements compared to standard RL on navigating a 7-DoF robot arm between obstacles.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Manipulation | scene-play oraclerep v0 | Task 1 Success Rate43 | 9 | |
| Manipulation | puzzle 3x3-play-oraclerep v0 | Task 1 Success Rate5 | 9 | |
| Manipulation | puzzle 4x4-play-oraclerep v0 | Task 1 Success Rate1 | 9 | |
| Navigation | pointmaze large-navigate-oraclerep v0 | Task 1 Success Rate40 | 9 | |
| Navigation | antmaze large navigate oraclerep v0 | Task 1 Success Rate6 | 9 | |
| Manipulation | cube-double-play oraclerep v0 | Task 1 Success Rate6 | 9 | |
| Manipulation | cube-single-play-oraclerep v0 | Task 1 Success Rate4 | 9 | |
| Navigation | humanoidmaze medium-navigate-oraclerep v0 | Task 1 Success Rate4 | 9 |