Impact of Feedback Type on Explanatory Interactive Learning

09/26/2022
by   Misgina Tsighe Hagos, et al.
0

Explanatory Interactive Learning (XIL) collects user feedback on visual model explanations to implement a Human-in-the-Loop (HITL) based interactive learning scenario. Different user feedback types will have different impacts on user experience and the cost associated with collecting feedback since different feedback types involve different levels of image annotation. Although XIL has been used to improve classification performance in multiple domains, the impact of different user feedback types on model performance and explanation accuracy is not well studied. To guide future XIL work we compare the effectiveness of two different user feedback types in image classification tasks: (1) instructing an algorithm to ignore certain spurious image features, and (2) instructing an algorithm to focus on certain valid image features. We use explanations from a Gradient-weighted Class Activation Mapping (GradCAM) based XIL model to support both feedback types. We show that identifying and annotating spurious image features that a model finds salient results in superior classification and explanation accuracy than user feedback that tells a model to focus on valid image features.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/12/2023

Learning from Exemplary Explanations

eXplanation Based Learning (XBL) is a form of Interactive Machine Learni...
research
03/04/2022

A Typology to Explore and Guide Explanatory Interactive Machine Learning

Recently, more and more eXplanatory Interactive machine Learning (XIL) m...
research
09/11/2023

Distance-Aware eXplanation Based Learning

eXplanation Based Learning (XBL) is an interactive learning approach tha...
research
10/30/2022

XMD: An End-to-End Framework for Interactive Explanation-Based Debugging of NLP Models

NLP models are susceptible to learning spurious biases (i.e., bugs) that...
research
09/07/2022

Semantic Interactive Learning for Text Classification: A Constructive Approach for Contextual Interactions

Interactive Machine Learning (IML) shall enable intelligent systems to i...
research
09/21/2021

SalienTrack: providing salient information for semi-automated self-tracking feedback with model explanations

Self-tracking can improve people's awareness of their unhealthy behavior...
research
09/08/2022

Improved Robust Algorithms for Learning with Discriminative Feature Feedback

Discriminative Feature Feedback is a setting proposed by Dastupta et al....

Please sign up or login with your details

Forgot password? Click here to reset