Visualizations of Deep Neural Networks in Computer Vision: A Survey

Seifert, Christin; Aamir, Aisha; Balagopalan, Aparna; Jain, Dhruv; Sharma, Abhinav; Grottel, Sebastian; Gumhold, Stefan

doi:10.1007/978-3-319-54024-5_6

Cited by 56 publications

(48 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given a black box solving a classification problem, the inspection problem consists in providing a representation for understanding either how the black box model works or why the black box returns certain predictions more likely than others. In [88], Seifet et al provide a survey of visualizations of DNNs by defining a classification scheme describing visualization goals and methods. They found that most papers use pixel displays to show neuron activations.…”

Section: Solving the Black Box Inspection Problemmentioning

confidence: 99%

A Survey of Methods for Explaining Black Box Models

et al. 2018

View full text Add to dashboard Cite

In the last years many accurate decision support systems have been constructed as black boxes, that is as systems that hide their internal logic to the user. This lack of explanation constitutes both a practical and an ethical issue. The literature reports many approaches aimed at overcoming this crucial weakness sometimes at the cost of scarifying accuracy for interpretability. The applications in which black box decision systems can be used are various, and each approach is typically developed to provide a solution for a specific problem and, as a consequence, delineating explicitly or implicitly its own definition of interpretability and explanation. The aim of this paper is to provide a classification of the main problems addressed in the literature with respect to the notion of explanation and the type of black box system. Given a problem definition, a black box type, and a desired explanation this survey should help the researcher to find the proposals more useful for his own work. The proposed classification of approaches to open black box models should also be useful for putting the many research open questions in perspective.

show abstract

Section: Solving the Black Box Inspection Problemmentioning

confidence: 99%

A Survey of Methods for Explaining Black Box Models

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Samek et al 42 Zhang and Zhu 39 Choo and Liu 30 Garcia et al 32 Grün et al 40 Hohman et al 20 Seifert et al 37 Yu and Shi 38 More general categories are represented in blue, while more specific ones are shown in red. DL seems to be the most popular subtopic of the surveys, with many papers describing the explanation of neural networks (NNs).…”

Section: Dudley and Kristensson 31mentioning

confidence: 99%

“…The two research questions that Seifert et al 37 tried to answer are ''What are the insights that can be gained from DNN models by using visualizations?'' and ''Which visualizations are appropriate for each kind of insights?''…”

mentioning

confidence: 99%

A survey of surveys on the use of visualization for interpreting machine learning models

Chatzimparmpas

Martins

Jusufi

et al. 2020

Information Visualization

113

View full text Add to dashboard Cite

Research in machine learning has become very popular in recent years, with many types of models proposed to comprehend and predict patterns and trends in data originating from different domains. As these models get more and more complex, it also becomes harder for users to assess and trust their results, since their internal operations are mostly hidden in black boxes. The interpretation of machine learning models is currently a hot topic in the information visualization community, with results showing that insights from machine learning models can lead to better predictions and improve the trustworthiness of the results. Due to this, multiple (and extensive) survey articles have been published recently trying to summarize the high number of original research papers published on the topic. But there is not always a clear definition of what these surveys cover, what is the overlap between them, which types of machine learning models they deal with, or what exactly is the scenario that the readers will find in each of them. In this article, we present a meta-analysis (i.e. a “survey of surveys”) of manually collected survey papers that refer to the visual interpretation of machine learning models, including the papers discussed in the selected surveys. The aim of our article is to serve both as a detailed summary and as a guide through this survey ecosystem by acquiring, cataloging, and presenting fundamental knowledge of the state of the art and research opportunities in the area. Our results confirm the increasing trend of interpreting machine learning with visualizations in the past years, and that visualization can assist in, for example, online training processes of deep learning models and enhancing trust into machine learning. However, the question of exactly how this assistance should take place is still considered as an open challenge of the visualization community.

show abstract

“…Explainable AI is currently a highly relevant research area that uses visualization to look into the black box that deep learning is often considered to be. We refer to Seifert et al [SAB*17], Hohman et al [HKPC18] and Ancona et al [ACOG18] for an introduction into this promising area.…”

Section: Related Workmentioning

confidence: 99%

Robust Reference Frame Extraction from Unsteady 2D Vector Fields with Convolutional Neural Networks

Kim

Günther

2019

Computer Graphics Forum

View full text Add to dashboard Cite

Robust feature extraction is an integral part of scientific visualization. In unsteady vector field analysis, researchers recently directed their attention towards the computation of near‐steady reference frames for vortex extraction, which is a numerically challenging endeavor. In this paper, we utilize a convolutional neural network to combine two steps of the visualization pipeline in an end‐to‐end manner: the filtering and the feature extraction. We use neural networks for the extraction of a steady reference frame for a given unsteady 2D vector field. By conditioning the neural network to noisy inputs and resampling artifacts, we obtain numerically stabler results than existing optimization‐based approaches. Supervised deep learning typically requires a large amount of training data. Thus, our second contribution is the creation of a vector field benchmark data set, which is generally useful for any local deep learning‐based feature extraction. Based on Vatistas velocity profile, we formulate a parametric vector field mixture model that we parameterize based on numerically‐computed example vector fields in near‐steady reference frames. Given the parametric model, we can efficiently synthesize thousands of vector fields that serve as input to our deep learning architecture. The proposed network is evaluated on an unseen numerical fluid flow simulation.

show abstract

Visualizations of Deep Neural Networks in Computer Vision: A Survey

Cited by 56 publications

References 47 publications

A Survey of Methods for Explaining Black Box Models

A Survey of Methods for Explaining Black Box Models

A survey of surveys on the use of visualization for interpreting machine learning models

Robust Reference Frame Extraction from Unsteady 2D Vector Fields with Convolutional Neural Networks

Contact Info

Product

Resources

About