Deep learning has deep problems

Stefan Mileschin · 17th January 2022, 10:47

Can’t multi-task

Deep learning has some major issues which will eventually lead it to become a dead end.

For 20 years, deep learning has come to dominate artificial intelligence research and applications through a series of useful commercial applications

But according to IEEE Spectrum the inability of a typical deep learning program to perform well on more than one task, for example, severely limits the application of the technology to specific tasks in rigidly controlled environments.

More seriously deep learning is untrustworthy because it is not explainable -- and unsuitable for some applications because it can experience catastrophic forgetting.

If the algorithm works, it may be impossible to fully understand why. And while the tool is slowly learning a new database, an arbitrary part of its learned memories can suddenly collapse. Therefore, it might be risky to use deep learning on any life-or-death application, such as a medical one.

https://fudzilla.com/news/ai/54205-d...-deep-problems

17th January 2022, 10:47	#1
Stefan Mileschin [M] Reviewer Join Date: May 2010 Location: Romania Posts: 148,618	Deep learning has deep problems Can’t multi-task Deep learning has some major issues which will eventually lead it to become a dead end. For 20 years, deep learning has come to dominate artificial intelligence research and applications through a series of useful commercial applications But according to IEEE Spectrum the inability of a typical deep learning program to perform well on more than one task, for example, severely limits the application of the technology to specific tasks in rigidly controlled environments. More seriously deep learning is untrustworthy because it is not explainable -- and unsuitable for some applications because it can experience catastrophic forgetting. If the algorithm works, it may be impossible to fully understand why. And while the tool is slowly learning a new database, an arbitrary part of its learned memories can suddenly collapse. Therefore, it might be risky to use deep learning on any life-or-death application, such as a medical one. https://fudzilla.com/news/ai/54205-d...-deep-problems

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Flex Logix solves Deep Learning DRAM problem	Stefan Mileschin	WebNews	0	2nd November 2018 11:53
Big Blue has a deep learning breakthough	Stefan Mileschin	WebNews	0	10th August 2017 06:31
Google can turn an ordinary PC into a deep learning machine	Stefan Mileschin	WebNews	0	22nd June 2017 05:31
Nvidia to train 100,000 developers on deep learning AI	Stefan Mileschin	WebNews	0	10th May 2017 05:50
Google’s deep learning computer can keep things secret	Stefan Mileschin	WebNews	0	31st October 2016 05:15
Microsoft makes its deep learning tools available to all	Stefan Mileschin	WebNews	0	26th October 2016 08:36
NVIDIA's insane DGX-1 is a computer tailor-made for deep learning	Stefan Mileschin	WebNews	0	7th April 2016 11:21
NVIDIA Launches World's First Deep Learning Supercomputer	Stefan Mileschin	WebNews	0	7th April 2016 11:20
MediaTek Helio X20 goes after deep learning	Stefan Mileschin	WebNews	0	2nd October 2015 06:42
NVIDIA Doubles Performance for Deep Learning Training	Stefan Mileschin	WebNews	0	9th July 2015 06:28