Program Highlights »
Tue Dec 6th 06:00 -- 09:30 PM @ Area 5+6+7+8 #178
Value Iteration Networks
Aviv Tamar · Sergey Levine · Pieter Abbeel · YI WU · Garrett Thomas

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.