Opening the Black Box: Revealing Interpretable Sequence Motifs in Kernel-Based Learning Algorithms

Vidovic, Marina M.-C.; Görnitz, Nico; Müller, Klaus‐Robert; Rätsch, Gunnar; Kloft, Marius

doi:10.1007/978-3-319-23525-7_9

Cited by 11 publications

(10 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the implicitly learned patterns, predictions, prescriptions, and classifications can be made. Research has attempted to understand black boxes by identifying the internal structures responsible for predictions (Vidovic et al, 2015). This is considered to be a first step towards interpreting implicit patterns.…”

Section: From Patterns To Theorymentioning

confidence: 99%

Data-Driven Meets Theory-Driven Research in the Era of Big Data: Opportunities and Challenges for Information Systems Research

Maaß¹,

Parsons²,

Purao³

et al. 2018

JAIS

View full text Add to dashboard Cite

The era of big data provides many opportunities for conducting impactful research from both datadriven and theory-driven perspectives. However, data-driven and theory-driven research have progressed somewhat independently. In this paper, we develop a framework that articulates important differences between these two perspectives and propose a role for information systems research at their intersection. The framework presents a set of pathways that combine the datadriven and theory-driven perspectives. From these pathways, we derive a set of challenges, and show how they can be addressed by research in information systems. By doing so, we identify an important role that information systems research can play in advancing both data-driven and theory-driven research in the era of big data. Medicine. His research investigates the relationship between conceptual modeling and AI/ML, AI in distributed environments (EdgeAI), and the design of smart services. Additionally he has a strong focus on knowledge transfer to industry through funded projects and spin-offs. 1273Carson Woo is Stanley Kwok Professor of Business, Sauder School of Business, University of British Columbia. His research interests include conceptual modeling, systems analysis and design, and requirements engineering. In particular, he is interested in using conceptual models to acquire relevant contextual information (e.g., business goals) and utilizing it to design new information systems, or aligning it to existing information systems design, so that changes can be more appropriate to business needs. At the University of British Columbia, he is a member of two research clusters: (1) Artificial Intelligence, and (2) Blockchain. Dr. Woo is editor of Information Technology and Systems Abstracts Journal at the Social Science Research Network, and serves or has served on several editorial boards, including ACM Transactions on Management Information Systems, Business & Information Systems Engineering Journal, Information and Management, and Requirements Engineering. He currently serves as the chair (2019-2020) of the Conceptual Modeling Conference steering committee and has served as the president of the

show abstract

Section: From Patterns To Theorymentioning

confidence: 99%

Data-Driven Meets Theory-Driven Research in the Era of Big Data: Opportunities and Challenges for Information Systems Research

Maaß¹,

Parsons²,

Purao³

et al. 2018

JAIS

View full text Add to dashboard Cite

show abstract

“…Technically, explanation methods have been applied to a broad range of models ranging from simple bag-ofwords-type classifiers or logistic regression [13], [28] to feedforward or recurrent deep neural networks [9], [11], [13], [171] and, more recently, also to unsupervised learning models [85], [86]. At the same time, these methods were able to handle different types of data, including images [13], speech [20], text [10], [38], and structured data, such as molecules [162], [165] or genetic sequences [191].…”

Section: S U C C E S S F U L U S E S O F E X P L a N A T I O N mentioning

confidence: 99%

Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications

Montavon²,

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Over the last few years much work has been done on "black box" model explanation. Some of this work (Adler et al, 2016;Baehrens et al, 2010;Lou et al, 2013;Montavon et al, 2018;Simonyan et al, 2014;Vidovic et al, 2015) has been aimed specifically at experts. The interpretability of a model is a key element of a robust validation procedure in applications such as medicine or self-driving cars.…”

Section: A Case Study: Building a Music Recommendation Systemmentioning

confidence: 99%

Black box algorithms and the rights of individuals: no easy solution to the “explainability” problem

Gryz¹,

Rojszczak²

2021

Internet Policy Review

View full text Add to dashboard Cite

Over the last few years, the interpretability of classification models has been a very active area of research. Recently, the concept of interpretability was given a more specific legal context. In 2016, the EU adopted the General Data Protection Regulation (GDPR), containing the right to explanation for people subjected to automated decision-making (ADM). The regulation itself is very reticent about what such a right might imply. As a result, since the introduction of the GDPR there has been an ongoing discussion about not only the need to introduce such a right, but also about its scope and practical consequences in the digital world. While there is no doubt that the right to explanation may be very difficult to implement due to technical challenges, any difficulty in explaining how algorithms work cannot be considered a sufficient reason to completely abandon this legal safeguard. The aim of this article is twofold. First, to demonstrate that the interpretability of "black box" machine learning algorithms is a challenging technical problem for which no solutions have been found. Second, to demonstrate how the explanation task should instead be completed using well-known and well-trialled IT solutions, such as event logging or statistical analysis of the algorithm. Based on the evidence exposed in this paper, the authors find that the most effective solution would be to benchmark the automated decision-making algorithms using certification frameworks, thus balancing the need to ensure adequate protection of individuals' rights with the understandable expectations of AI technology providers to have their intellectual property rights protected.

show abstract

Opening the Black Box: Revealing Interpretable Sequence Motifs in Kernel-Based Learning Algorithms

Cited by 11 publications

References 26 publications

Data-Driven Meets Theory-Driven Research in the Era of Big Data: Opportunities and Challenges for Information Systems Research

Data-Driven Meets Theory-Driven Research in the Era of Big Data: Opportunities and Challenges for Information Systems Research

Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications

Black box algorithms and the rights of individuals: no easy solution to the “explainability” problem

Contact Info

Product

Resources

About