Attention-Based Prototypical Learning Towards Interpretable, Confident and Robust Deep Neural Networks

02/17/2019
by   Sercan O. Arik, et al.
4

We propose a new framework for prototypical learning that bases decision-making on few relevant examples that we call prototypes. Our framework utilizes an attention mechanism that relates the encoded representations to determine the prototypes. This results in a model that: (1) enables interpretability by outputting samples most relevant to the decision-making in addition to outputting the classification results; (2) allows confidence-controlled prediction by quantifying the mismatch across prototype labels; (3) permits detection of distribution mismatch; and (4) improves robustness to label noise. We demonstrate that our model is able to maintain comparable performance to baseline models while enabling all these benefits.

READ FULL TEXT

page 7

page 14

research
08/20/2019

TabNet: Attentive Interpretable Tabular Learning

We propose a novel high-performance interpretable deep tabular data lear...
research
10/15/2022

DProtoNet: Decoupling the inference module and the explanation module enables neural networks to have better accuracy and interpretability

The interpretation of decisions made by neural networks is the focus of ...
research
08/13/2021

TDM: Trustworthy Decision-Making via Interpretability Enhancement

Human-robot interactive decision-making is increasingly becoming ubiquit...
research
09/07/2020

Attention based Writer Independent Handwriting Verification

The task of writer verification is to provide a likelihood score for whe...
research
03/14/2022

A Novel Perspective to Look At Attention: Bi-level Attention-based Explainable Topic Modeling for News Classification

Many recent deep learning-based solutions have widely adopted the attent...
research
03/27/2020

A copula-based visualization technique for a neural network

Interpretability of machine learning is defined as the extent to which h...
research
08/26/2018

Semantic-Unit-Based Dilated Convolution for Multi-Label Text Classification

We propose a novel model for multi-label text classification, which is b...

Please sign up or login with your details

Forgot password? Click here to reset