# Guided backprop interpretation for multilayer perceptron

**URL:** <https://discuss.pytorch.org/t/guided-backprop-interpretation-for-multilayer-perceptron/170765>\
**Category:** Captum\
**Created:** [January 21, 2023, 4:49am UTC](https://discuss.pytorch.org/t/guided-backprop-interpretation-for-multilayer-perceptron/170765 "2023-01-21T04:49:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![neurospacer](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/neurospacer/32/47962_2.png) [@neurospacer](https://discuss.pytorch.org/u/neurospacer)\
**Post date:** [January 21, 2023, 4:49am UTC](https://discuss.pytorch.org/t/guided-backprop-interpretation-for-multilayer-perceptron/170765/1 "2023-01-21T04:49:25Z")

</div>

Hi,

I want to use interpretation algorithm for simple feed-forward neural network (multilayer perceptron with 13 input neurons, 2 layers deep - 10 neurons each, output is 2 class) to classify voxels.

Each input neuron comes as voxel from 13 feature maps channels of the CT image (extracted statistical maps). The question is can I use [Guided Backpropagation](https://captum.ai/api/guided_backprop.html) for interpretation? Or it better to use model-agnostic methods like LIME?

I didn’t find any use cases of guided backpropagation for MLP, most of the examples are used with deep convolutional NNs.