[Submitted on 7 May 2020]
Abstract: In this paper we introduce plan2vec, an unsupervised representation learning
approach that is inspired by reinforcement learning. Plan2vec constructs a
weighted graph on an image dataset using near-neighbor distances, and then
extrapolates this local metric to a global embedding by distilling
path-integral over planned path. When applied to control, plan2vec offers a way
to learn goal-conditioned value estimates that are accurate over long horizons
that is both compute and sample efficient. We demonstrate the effectiveness of
plan2vec on one simulated and two challenging real-world image datasets.
Experimental results show that plan2vec successfully amortizes the planning
cost, enabling reactive planning that is linear in memory and computation
complexity rather than exhaustive over the entire state space.
Submission history
From: Ge Yang [view email]
[v1]
Thu, 7 May 2020 17:52:23 UTC (5,786 KB)
