GANS for Sequences of Discrete Elements with the Gumbel-softmax Distribution
Kusner, Hernández-Lobato · stat.ML,cs.LG · 2016-11-12 · 原文
Generative Adversarial Networks (GAN) have limitations when the goal is to generate sequences of discrete elements. The reason for this is that samples from a distribution on discrete objects such as the multinomial are not differentiable with respect to the distribution parameters. This problem can be avoided by using the Gumbel-softmax distribution, which is a continuous approximation to a multinomial distribution parameterized in terms of the softmax function. In this work, we evaluate the performance of GANs based on recurrent neural networks with Gumbel-softmax output distributions in the task of generating sequences of discrete elements.
讲义
讲义·推断 依据「原文」自动生成的结构化摘要（推断），非原文表述；以原文为准。
1. 人话版
Generative Adversarial Networks (GAN) have limitations when the goal is to generate sequences of discrete elements.
The reason for this is that samples from a distribution on discrete objects such as the multinomial are not differentiable with respect to the distribution parameters.
2. 领域脉络
本文类目：stat.ML、cs.LG，属于其所在研究脉络的最新进展。
3. 机制拆解
摘要未展开方法细节——精读时重点看方法/模型部分。
4. 证据与数字
摘要未给出量化结果——留意原文的实验与数据。
5. 反例与边界
This problem can be avoided by using the Gumbel-softmax distribution, which is a continuous approximation to a multinomial distribution parameterized in terms of the softmax function.
In this work, we evaluate the performance of GANs based on recurrent neural networks with Gumbel-softmax output distributions in the task of generating sequences of discrete elements.
6. 跨领域连接与意外收获
横跨 2 个类目（stat.ML、cs.LG），关注其在你兴趣板块间的迁移面。
7. 可复用方法
把本文机制与你手头项目对照，找一个两周内能验证的最小实验。
8. 术语表
精读时把不熟的术语记入此处，作为下次回忆的锚点。