02556naa a22003258i 4500 5315 5315 00005315 FR-PaOEC 20211006062616.0 a o d i cr || |||m|n|| 190601s2017 ||| o i|0| 0 eng d (FR-PaOEC) FR-PaOEC ES-MaONT Samek, Wojciech 3410 Explainable artificial intelligence : Understanding, visualizing and interpreting deep learning models / Wojciech Samek, Thomas Wiegand and Klaus-Robert Müller Geneva : International Telecommunication Union, 2017. 10 p. texto (visual) isbdcontent electrónico isbdmedia recurso en línea rdacarrier With the availability of large databases and recent improvements in deep learning methodology, the performance of AI systems is reaching, or even exceeding, the human level on an increasing number of complex tasks. Impressive examples of this development can be found in domains such as image classification, sentiment analysis, speech understanding or strategic game playing. However, because of their nested non-linear structure, these highly successful machine learning and artificial intelligence models are usually applied in a black-box manner, i.e. no information is provided about what exactly makes them arrive at their predictions. Since this lack of transparency can be a major drawback, e.g. in medical applications, the development of methods for visualizing, explaining and interpreting deep learning models has recently attracted increasing attention. This paper summarizes recent developments in this field and makes a plea for more interpretability in artificial intelligence. Furthermore, it presents two approaches to explaining predictions of deep learning models, one method which computes the sensitivity of the prediction with respect to changes in the input and one approach which meaningfully decomposes the decision in terms of the input variables. These methods are evaluated on three classification tasks. Tecnologías habilitadoras digitales 18 Science and Technology Wiegand, Thomas 3413 Müller, Klaus-Robert 3330 4843 ITU Journal: ICT Discoveries Vol. 2018, no. 1, p. 39-48 2018:1<39 2616-8375 itu-ilibrary.org https://www.itu.int/dms_pub/itu-s/opb/journal/S-JOURNAL-ICTS.V1I1-2017-5-PDF-E.pdf Acceso al documento 0 pdf ART udc 0 0 0 3 1 6265 01 01 online 2019-09-12 0 1000020175564 2019-09-12 00:00:00 2019-09-12 ART pdf