arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1905.06982cs.LGstat.ML
收藏

适用于可变大小输入的高效深度高斯过程模型

Efficient Deep Gaussian Process Models for Variable-Sized Input

Element AI;University of British Columbia(不列颠哥伦比亚大学);Rutgers University(罗格斯大学);Samsung AI Center(三星人工智能中心)

Issam H. Laradji, Mark Schmidt, Vladimir Pavlovic, Minyoung Kim

摘要:

深度高斯过程(DGP)具有吸引人的贝叶斯属性,可以处理可变大小的数据,并学习深度特征。它们的局限性在于它们不能很好地随数据的大小而扩展。现有的方法使用深度随机特征(DRF)扩展模型来解决这个问题,该模型通过近似DGP来使推理易于处理。然而,DRF不适合于可变大小的输入数据,例如树、图和序列。我们介绍了GP-DRF,一种新的贝叶斯模型与输入层的GP,其次是DRF层。GP和DRF的结合产生了一个易于处理的模型,既可以处理可变大小的输入,也可以学习数据的深度长期依赖结构。我们提供了一种新的有效的方法,同时推断GP的潜向量的后验和推断DRF的内部权重和随机频率的后验。我们的实验表明,GP-DRF在许多数据集上优于标准GP模型和DRF模型。此外,他们证明,GP-DRF能够改善不确定性量化相比,GP和DRF单独,相对于巴塔查里亚距离评估。源代码可在https://github.com/IssamLaradji/GP_DRF上获得。

英文摘要:

Deep Gaussian processes (DGP) have appealing Bayesian properties, can handle variable-sized data, and learn deep features. Their limitation is that they do not scale well with the size of the data. Existing approaches address this using a deep random feature (DRF) expansion model, which makes inference tractable by approximating DGPs. However, DRF is not suitable for variable-sized input data such as trees, graphs, and sequences. We introduce the GP-DRF, a novel Bayesian model with an input layer of GPs, followed by DRF layers. The key advantage is that the combination of GP and DRF leads to a tractable model that can both handle a variable-sized input as well as learn deep long-range dependency structures of the data. We provide a novel efficient method to simultaneously infer the posterior of GP's latent vectors and infer the posterior of DRF's internal weights and random frequencies. Our experiments show that GP-DRF outperforms the standard GP model and DRF model across many datasets. Furthermore, they demonstrate that GP-DRF enables improved uncertainty quantification compared to GP and DRF alone, with respect to a Bhattacharyya distance assessment. Source code is available at https://github.com/IssamLaradji/GP_DRF.

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。

↑