Sökning: "Förklaringsbarhet"
Hittade 2 uppsatser innehållade ordet Förklaringsbarhet.
1. Investigating the Use of Deep Learning Models for Transactional Underwriting
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Tabular data is the most common form of data, and is abundant throughout crucial industries, such as banks, hospitals and insurance companies. Albeit, deep learning research has largely been dominated by applications to homogeneous data, e.g. images or natural language. LÄS MER
2. Explainable AI - Visualization of Neuron Functionality in Recurrent Neural Networks for Text Prediction
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Artificial Neural Networks are successfully solving a wide range of problems with impressive performance. Nevertheless, often very little or nothing is understood in the workings behind these black-box solutions as they are hard to interpret, let alone to explain. LÄS MER