The goal of Interpretable Machine Learning is to allow oversight and understanding of machine-learned decisions. Much of the work in Interpretable Machine Learning has come in the form of devising methods to better explain the predictions of machine learning models.
NAMs learn a linear combination of neural networks that each attend to a single input feature.
Multi-horizon forecasting problems often contain a complex mix of inputs -- including static (i. e. time-invariant) covariates, known future inputs, and other exogenous time series that are only observed historically -- without any prior information on how they interact with the target.
We propose a novel inherently interpretable machine learning method that bases decisions on few relevant examples that we call prototypes.
Understanding why a model makes a certain prediction can be as crucial as the prediction's accuracy in many applications.
Here we present DeepLIFT (Deep Learning Important FeaTures), a method for decomposing the output prediction of a neural network on a specific input by backpropagating the contributions of all neurons in the network to every feature of the input.
Despite widespread adoption, machine learning models remain mostly black boxes.
The first is a tool that visualizes the activations produced on each layer of a trained convnet as it processes an image or video (e. g. a live webcam stream).
The presented library iNNvestigate addresses this by providing a common interface and out-of-the- box implementation for many analysis methods, including the reference implementation for PatternNet and PatternAttribution as well as for LRP-methods.