Redirigiendo al acceso original de articulo en 15 segundos...
Inicio  /  Algorithms  /  Vol: 15 Par: 10 (2022)  /  Artículo
ARTÍCULO
TITULO

Comparing Approaches for Explaining DNN-Based Facial Expression Classifications

Kaya ter Burg and Heysem Kaya    

Resumen

Classifying facial expressions is a vital part of developing systems capable of aptly interacting with users. In this field, the use of deep-learning models has become the standard. However, the inner workings of these models are unintelligible, which is an important issue when deploying them to high-stakes environments. Recent efforts to generate explanations for emotion classification systems have been focused on this type of models. In this work, an alternative way of explaining the decisions of a more conventional model based on geometric features is presented. We develop a geometric-features-based deep neural network (DNN) and a convolutional neural network (CNN). Ensuring a sufficient level of predictive accuracy, we analyze explainability using both objective quantitative criteria and a user study. Results indicate that the fidelity and accuracy scores of the explanations approximate the DNN well. From the performed user study, it becomes clear that the explanations increase the understanding of the DNN and that they are preferred over the explanations for the CNN, which are more commonly used. All scripts used in the study are publicly available.

 Artículos similares

       
 
Anna Marino, Francesca Bona, Stefano Fenoglio and Tiziano Bo    
Of all the environmental elements that influence the biological communities of rivers, water flow characteristics are undoubtedly the most important. Unfortunately, natural hydrological characteristics are increasingly threatened by human activities, esp... ver más
Revista: Water

 
Xiaobang Wang, Yang Yu, Siyu Li, Jie Zhang and Zhijie Liu    
The Revolving Floating Crane (RFC) is a specialized engineering vessel crucial for offshore lifting operations, such as offshore platform construction and deep-water salvaging. It boasts impressive lifting capacity, good adaptability to various environme... ver más

 
Michiel van der Vlag, Lionel Kusch, Alain Destexhe, Viktor Jirsa, Sandra Diaz-Pier and Jennifer S. Goldman    
Global neural dynamics emerge from multi-scale brain structures, with nodes dynamically communicating to form transient ensembles that may represent neural information. Neural activity can be measured empirically at scales spanning proteins and subcellul... ver más
Revista: Applied Sciences

 
Alessia Di Giovanni, Diego Di Curzio, Davide Pantanella, Cristiana Picchi and Sergio Rusi    
Recently, new numerical methods have been applied to weather data for the estimation of water budget, especially when the lack of measured data is considerable. Geostatistics is one of the most powerful approaches when it comes to studying spatially rele... ver más
Revista: Water

 
Huashuai Liu, Fan Yang and Hongchuan Wang    
Climate change poses higher requirements on ocean engineering design, and reasonable estimation of design wave heights plays a crucial role in coastal protection and offshore engineering. Extreme value analysis is widely used in frequency calculations of... ver más
Revista: Water