1: \begin{abstract}

2: Significance testing aims to determine whether a proposition about the population distribution is the truth or not given observations.

3: However, traditional significance testing often needs to derive the distribution of the testing statistic, failing to deal with complex nonlinear relationships.

4: In this paper, we propose to conduct Full Bayesian Significance Testing for neural networks, called \textit{n}FBST, to overcome the limitation in relationship characterization of traditional approaches.

5: A Bayesian neural network is utilized to fit the nonlinear and multi-dimensional relationships with small errors and avoid hard theoretical derivation by computing the evidence value.

6: % , a function of the posterior distributions of the null hypothesis and the alternative.

7: %, with a theoretical convergence.

8: %\textbf{We also theoretically prove that the evidence value of a sharp null hypothesis goes to one as the sample size goes to infinite.}

9: Besides, \textit{n}FBST can test not only global significance but also local and instance-wise significance, which previous testing methods don't focus on.

10: Moreover, \textit{n}FBST is a general framework that can be extended based on the measures selected, such as Grad-\textit{n}FBST, LRP-\textit{n}FBST, DeepLIFT-\textit{n}FBST, LIME-\textit{n}FBST. 

11: A range of experiments on both simulated and real data are conducted to show the advantages of our method. 

12: \end{abstract}

13: