Date: Fri, 05 Jun 2020 19:59:00 +0000
<div class="container"> <p>Deep neural networks are undeniably effective. They rely on such a high number of parameters, that they are appropriately described as “black boxes”.</p> <p>While black boxes lack desirably properties like interpretability and explainability, in some cases, their accuracy makes them incredibly useful.</p> <p>But does achiving “usefulness” require a black box? Can we be sure an equally valid but simpler solution does not exist?</p> <p><a href="https://users.cs.duke.edu/~cynthia/">Cynthia Rudin</a> helps us answer that question. We discuss her recent paper with co-author Joanna Radin titled (spoiler warning)…</p> <p><a href="https://hdsr.mitpress.mit.edu/pub/f9kuryi8/release/5">Why Are We Using Black Box Models in AI When We Don’t Need To? A Lesson From An Explainable AI Competition</a></p> </div> <p><br /> <br /> <br /></p>
