Value Iteration Networks

We introduce the value iteration network (VIN): a fully differentiable neural network with a `planning module' embedded within. VINs can learn to plan, and are suitable for predicting outcomes that involve planning-based reasoning, such as policies for reinforcement learning. Key to our approach is a novel differentiable approximation of the value-iteration algorithm, which can be represented as a convolutional neural network, and trained end-to-end using standard backpropagation. We evaluate VIN based policies on discrete and continuous path-planning domains, and on a natural-language based search task. We show that by learning an explicit planning computation, VIN policies generalize better to new, unseen domains. This paper is a significantly abridged and IJCAI audience targeted version of the original NIPS 2016 paper with the same title, available here: https: //arxiv. org/abs/1602. 02867

Authors

• Aviv Tamar
• Yi Wu Tsinghua University Shanghai Qi Zhi Institute
• Garrett Thomas
• Sergey Levine
• Pieter Abbeel UC Berkeley

Keywords

• Artificial Intelligence: artificial intelligence
• Artificial Intelligence: automated planning