DeepAI AI Chat
Log In Sign Up

Interpretable Neural Predictions with Differentiable Binary Variables

by   Joost Bastings, et al.
University of Amsterdam

The success of neural networks comes hand in hand with a desire for more interpretability. We focus on text classifiers and make them more interpretable by having them provide a justification, a rationale, for their predictions. We approach this problem by jointly training two neural network models: a latent model that selects a rationale (i.e. a short and informative part of the input text), and a classifier that learns from the words in the rationale alone. Previous work proposed to assign binary latent masks to input positions and to promote short selections via sparsity-inducing penalties such as L0 regularisation. We propose a latent model that mixes discrete and continuous behaviour allowing at the same time for binary selections and gradient-based training without REINFORCE. In our formulation, we can tractably compute the expected value of penalties such as L0, which allows us to directly optimise the model towards a pre-specified text selection rate. We show that our approach is competitive with previous work on rationale extraction, and explore further uses in attention mechanisms.


Learning Variational Word Masks to Improve the Interpretability of Neural Text Classifiers

To build an interpretable neural text classifier, most of the prior work...

SPECTRA: Sparse Structured Text Rationalization

Selective rationalization aims to produce decisions along with rationale...

DoLFIn: Distributions over Latent Features for Interpretability

Interpreting the inner workings of neural models is a key step in ensuri...

Training Competitive Binary Neural Networks from Scratch

Convolutional neural networks have achieved astonishing results in diffe...

Latent Transformations for Discrete-Data Normalising Flows

Normalising flows (NFs) for discrete data are challenging because parame...

Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery

The strength of modern generative models lies in their ability to be con...

Multi-Dimensional Explanation of Reviews

Neural models achieved considerable improvement for many natural languag...