arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1606.05908stat.MLcs.LG
收藏

可变自编码器

Tutorial on Variational Autoencoders

Carnegie Mellon(卡内基梅隆大学);UC Berkeley(加州大学伯克利分校)

Carl Doersch

摘要:

在短短三年的时间里,变分自编码器(VAE)已经成为复杂分布无监督学习的最流行方法之一。VAE之所以吸引人,是因为它们建立在标准函数逼近器(神经网络)之上,并且可以用随机梯度下降进行训练。VAE已经在生成许多类型的复杂数据方面显示出了希望,包括手写数字、人脸、门牌号、CIFAR图像、场景的物理模型、分割以及从静态图像预测未来。本教程介绍VAE背后的直觉,解释它们背后的数学,并描述一些经验行为。假设变分贝叶斯方法没有先验知识。

英文摘要:

In just three years, Variational Autoencoders (VAEs) have emerged as one of the most popular approaches to unsupervised learning of complicated distributions. VAEs are appealing because they are built on top of standard function approximators (neural networks), and can be trained with stochastic gradient descent. VAEs have already shown promise in generating many kinds of complicated data, including handwritten digits, faces, house numbers, CIFAR images, physical models of scenes, segmentation, and predicting the future from static images. This tutorial introduces the intuitions behind VAEs, explains the mathematics behind them, and describes some empirical behavior. No prior knowledge of variational Bayesian methods is assumed.

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。

↑