BrambleXu / knowledge-graph-learning

A curated list of awesome knowledge graph tutorials, projects and communities.
MIT License
736 stars 120 forks source link

arXiv-2019/06-Weight Agnostic Neural Networks #234

Open BrambleXu opened 5 years ago

BrambleXu commented 5 years ago

一句话总结:

主要是针对强化学习这个背景下的研究。不同的神经网络结构,有些结构在特定任务上,就是会比其他任务效果要好。(和NLP关系不是很大,但是挺有意思的)

资源:

论文信息:

笔记:

In this work, we develop neural networks with architectures that are naturally capable of performing a given task even when its weight parameters are randomly sampled.

提出了一个方法,用来判断简单的神经网络架构,在一个特定的task上,会有inductive biases.

模型图:

结果

接下来要看的论文