Computer Vision News - May 2021

3 Spotlight News 3 Hear How AI Software Wrote a ‘New’ Nirvana Song We head now even further West, to Toronto . I am sure that many of you will like this one! An organization has created a “new” Nirvana song using AI software to imitate Kurt Cobain’s songwriting. The music and lyrics bear evocative, Cobain-esque qualities which you will undoubtedly recognize. Band frontman Eric Hogan and his teammates say that nearly everything on the song is the work of computers. The intention of the organizers Lost Tapes of the 27 Club is to draw attention to the tragedy of death by suicide and how living musicians can get help with depression. Don’t miss the video at the right! Facebook’s Rotational AI Science & Engineering Program (RAISE) To conclude this month’s AI Spotlight News, let us tell you of this info session about Facebook’s new RAISE Program . Register now because the session will be on May 4 , just one day after the publication of this magazine: no time to waste. We only tell you that RAISE is a 24-month program offering participants full-time employment within Facebook AI , with an open option for software engineers to integrate a permanent Facebook AI team that aligns with their skills and interests . The impressive roster of speakers includes great friends of Computer Vision News Tamara Berg and Cristian Canton . Read More Why Computers Will Likely Never Perform Abductive Inferences Philosopher and programmer Erik Larson wrote a book called The Myth of Artificial Intelligence . We have liked this review by mathematician and philosopher William A. Dembski . It says that the key to understanding why computers cannot, and most likely will never, be able to perform abductive inferences is the problem of underdetermination of explanation by data. To get the computer to achieve anything, you need to input a massive amount of data, which needs to be painstakingly programmed. Differently from humans, who can get to the right conclusion about many things on the basis of incomplete data. Read More

RkJQdWJsaXNoZXIy NTc3NzU=