{"id":1559,"date":"2022-02-16T14:29:15","date_gmt":"2022-02-16T14:29:15","guid":{"rendered":"https:\/\/salarydistribution.com\/machine-learning\/2022\/02\/16\/the-greatest-podcast-ever-recorded\/"},"modified":"2022-02-16T14:29:15","modified_gmt":"2022-02-16T14:29:15","slug":"the-greatest-podcast-ever-recorded","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2022\/02\/16\/the-greatest-podcast-ever-recorded\/","title":{"rendered":"The Greatest Podcast Ever Recorded"},"content":{"rendered":"
\n

Is this the best podcast ever recorded? Let\u2019s just say you don\u2019t need a GPU to know that\u2019s a stretch. But it\u2019s pretty great if you\u2019re a fan of tall tales.<\/p>\n

And better still if you\u2019re not a fan of stretching the truth at all.<\/p>\n

That\u2019s because detecting hyperbole may one day get more manageable, thanks to researchers at the University of Copenhagen working in the growing field of exaggeration detection.<\/p>\n

Dustin Wright and Isabelle Augenstein have used NVIDIA GPUs to train an \u201cexaggeration detection system\u201d to identify overenthusiastic claims in health science reporting.<\/p>\n

Their work comes as the pandemic has fueled demand for understandable, accurate information. And social media has made health misinformation more widespread.<\/p>\n

Their paper leverages \u201cfew-shot learning,\u201d a technique that lets developers wring more intelligence out of less data, and a new version of a technique called pattern exploiting training.<\/p>\n

Research like Wright and Augenstein\u2019s could one day speed more precise health sciences news to more people.<\/p>\n

AI Podcast host Noah Kravitz \u2014 whose fishing stories we will never trust again after this episode \u2014 spoke with Wright about the work.<\/p>\n

Key Points From This Episode<\/b><\/h2>\n