arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1602.02867cs.AIcs.LGcs.NEstat.ML
收藏

价值迭代网络

Value Iteration Networks

UC Berkeley(加州大学伯克利分校)

Aviv Tamar, Yi Wu, Garrett Thomas, Sergey Levine, Pieter Abbeel

摘要:

我们介绍价值迭代网络(VIN):一个完全可微的神经网络与“规划模块”嵌入其中。VIN可以学习计划,并且适合预测涉及基于计划的推理的结果,例如强化学习的策略。我们的方法的关键是一种新的可微近似值迭代算法,它可以表示为一个卷积神经网络,并使用标准的反向传播训练端到端。我们评估VIN基于离散和连续路径规划域的政策,并在自然语言为基础的搜索任务。我们表明,通过学习一个明确的规划计算,VIN政策更好地推广到新的,看不见的领域。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑