Peering inside an Al's brain will help us trust it

• Autores: Matt Reynolds
• Localización: New scientist, ISSN 0262-4079, Nº. 3133, 2017, pág. 10
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Chris Grimm at Brown University in Providence and his colleagues created a system that analyzes which part of an image an AI is focusing on when it decides what the image is depicting. Similarly, for a document-sorting algorithm, the system highlights which words the algorithm used to decide which category a particular document should belong to. It's really useful to be able to look at an AI and find out how it's learning, says Dumitru Erhan at Google. Grimm's tool provides a handy way to check an algorithm is coming up with the right answer for the right reasons, he says.