Differentiable Particle Filters: End-to-End Learning with Algorithmic Priors
Differentiable Particle Filters: End-to-End Learning with Algorithmic Priors
复制标题
DOI：
10.15607/rss.2018.xiv.001
复制
发表时间：
2018-05
期刊：
影响因子：
--
通讯作者：
Rico Jonschkowski;Divyam Rastogi;O. Brock
中科院分区：
文献类型：
--
作者：
Rico Jonschkowski;Divyam Rastogi;O. Brock
我们提出了可微粒子过滤器(DPFS)：一种粒子滤波算法的可微实现，具有可学习的运动和测量模型。由于DPF是端到端可微的，我们可以通过优化端到端的状态估计性能来有效地训练它们的模型，而不是像模型精度这样的代理目标。DPF利用对状态上的概率分布进行操作的预测和测量更新来编码递归状态估计的结构。该结构表示一种算法先验，该算法先验改进了状态估计问题中的学习性能，同时允许学习模型的可解释性。我们在模拟和真实数据上的实验表明，端到端的算法先验学习具有很大的好处，例如将错误率降低了约80%。我们的实验还表明，与长时间的短期记忆网络不同，DPF以一种与策略无关的方式学习本地化，从而极大地提高了泛化能力。源代码可在此HTTPS URL上找到。
We present differentiable particle filters (DPFs): a differentiable implementation of the particle filter algorithm with learnable motion and measurement models. Since DPFs are end-to-end differentiable, we can efficiently train their models by optimizing end-to-end state estimation performance, rather than proxy objectives such as model accuracy. DPFs encode the structure of recursive state estimation with prediction and measurement update that operate on a probability distribution over states. This structure represents an algorithmic prior that improves learning performance in state estimation problems while enabling explainability of the learned model. Our experiments on simulated and real data show substantial benefits from end-to- end learning with algorithmic priors, e.g. reducing error rates by ~80%. Our experiments also show that, unlike long short-term memory networks, DPFs learn localization in a policy-agnostic way and thus greatly improve generalization. Source code is available at this https URL .