Trending Topics

  1. Real VS pretentious data scientists Troy Sadkowsky 01-Mar-2017

Latest News from DS

From Bhutan 19-Feb-2017

Can we unlock the Deep Learning black box?

Troy Sadkowsky - Wednesday, June 28, 2017

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Times; color: #000000; -webkit-text-stroke: #000000} p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Times; color: #000000; -webkit-text-stroke: #000000; min-height: 14.0px} span.s1 {font-kerning: none} span.s2 {text-decoration: underline ; font-kerning: none; color: #0000ff; -webkit-text-stroke: 0px #0000ff}

Can we unlock the Deep Learning black box?

Artificial intelligence has experienced a boom in recent years due to the increasing automation and generation of big data. Due to deep learning and artificial neural networks, machines are acquiring new perceptual abilities, such as recognizing images, speech, or reading handwriting, paving the way for an infinite number of new applications in human lives. However, this leap forward in AI processing comes at a cost – the reasoning behind the choices that deep learning networks make has become inscrutable to the engineers that built them.

Progress in automated vision greatly benefits from deep learning because general visual processing is far too complex to code by hand. A program that can learn by observing human example or through complicated training sets, and can generate its own algorithms through an interconnected network of a dozen to several hundred layers of simulated neurons, positions us that much closer to a self-driving car, or other AI programs capable of sophisticated automated decision-making. But if we cannot understand a program’s individual decisions – why it chose to drive into a tree, why a patient’s medications should be changed, why one individual was hired for the job and another flagged as a terrorist – what ethical and moral boundaries are we crossing? And if an artificial intelligence program is not 100% accurate, as they rarely are, what are the dangers of putting ourselves at the mercy of a machine’s mistakes?

According to an article in MIT Technology Review (https://www.technologyreview.com/s/604087/the-dark-secret-at-the-heart-of-ai/), trust will be a key factor in future applications of deep learning. Once machines can explain their reasoning to humans, we will better be able to learn from and act on their insights. Research is underway to develop tools for machine learning programs to be able to explain themselves. Regina Barzilay, at MIT, is developing a system that can collaborate with doctors by extracting snippets of text that represent patterns it has discover, and Carlos Guestrin’s system, from the University of Washington, highlights significant keywords or parts of an image to support a particular decision.

However, interpretable deep learning is far off, and even then, explanations offered by machines will always be simplified to a degree, which will always make trusting machine learning programs controversial. As a consequence, the European Union might soon pass legislation to make explanations for automated decisions a fundamental legal right, required of the companies that implement AI machine programs.

Jeff Clune, an investigator at the University of Wyoming testing deep neural networks, offers that, just as with human intelligence and decision-making, it is perhaps “the nature of intelligence that only part of it is exposed to rational explanation. Some of it is just instinctual, or subconscious, or inscrutable”. As deep learning offers more appealing possibilities in medicine, technology and other industries, when and should, society take the deep learning leap of faith?

Comments
Quentin commented on 09-Nov-2018 05:21 AM
Wow, gorgeous portal. Thnx ...

Here is my website;FIFA55
free android scales commented on 09-Nov-2018 05:49 AM
I blog quite often and I really thank you for your information. The article
has truly peaked my interest. I'm going to take a note of your blog and keep checking for new
details about once per week. I opted in for your Feed as well.
wanelo.co commented on 09-Nov-2018 07:27 AM
Everything is very open with a really clear description of the challenges.
It was definitely informative. Your website is very useful.
Many thanks for sharing!
Felica commented on 09-Nov-2018 09:05 AM
I like this site - its so usefull and helpfull.


My web siteQuest Bars Cheap
188bet commented on 09-Nov-2018 10:25 AM
This work reveals some sort of poetic mood and everyone would easily be attracted by it.
If this is a matter of yours too, then you definitely should learn concerning the best techniques
to procure such things. The beginning of Leonardo's life was
dedicated to art and painting in particular.
agen bola commented on 09-Nov-2018 01:00 PM
Appreciate this post. Will try it out.
바카라사이트 commented on 09-Nov-2018 01:33 PM
Aw, this was an extremely good post. Taking the time and actual effort to
create a great article? but what can I say?
I put things off a lot and never seem to get nearly anything done. https://www.banker9.net
stays on topic and states valid points. Thank you.
188bet commented on 09-Nov-2018 02:21 PM
Should your motive the following is to learn paintings available for sale Melbourne or paintings available Brisbane, unfortunately however, you can't view it here.

After the Bourbon Restoration, because trial participant of Louis XVI, David
was deprived of his civil right and property, and
was made to leave his homeland to stay in Brussels where David also completed many works, and lastly died in the strange land.
Then it is not important whether it is heads or tail,
one can predict a final results.
Francesca commented on 09-Nov-2018 05:30 PM
Hi there, well put together internet site you have got here.


my page:Can anybody use PLR articles for writing an eBook?
Juliane commented on 09-Nov-2018 05:49 PM
I really like what you guys are usually up too. This kind of clever work and coverage!
Keep up the wonderful works guys I've included you guys to my personal blogroll.

Also visit my site;Personal Struggle
https://github.com/jasonhocky/vincity/issues/1 commented on 09-Nov-2018 06:37 PM
I'm now not sure where you're getting your information, however good topic.
I needs to spend a while finding out much more or figuring
out more. Thanks for excellent info I used to be searching for this info
for my mission.
Your site has proven useful to me.
Teddy commented on 09-Nov-2018 08:37 PM
This excellent website really has all of the information I wanted concerning this subject and didn�t know who to ask.



My webpage;Https://Kumpulanitu.com
shopping online commented on 09-Nov-2018 10:40 PM
Hi there, I found your website via Google even as searching for a similar matter, your
web site came up, it seems to be great. I've bookmarked it in my google bookmarks.

Hello there, simply was aware of your blog through Google,
and located that it is truly informative.
I'm gonna be careful for brussels. I'll be grateful when you continue this in future.
Lots of people will be benefited out of your writing.
Cheers!
keo nha cai commented on 10-Nov-2018 12:39 AM
Should your motive here is to find out paintings available Melbourne or
paintings for sale Brisbane, unfortunately nevertheless, you can't see it here.
Waterslide paper emerged in clear or white however clear is more
preferred, given that almost any unprinted locations around the
image remains clear. Then it does not matter if it is heads or tail, it's possible to predict the ultimate results.
Shayne commented on 10-Nov-2018 06:52 AM
Wow cuz this What Equipment is Needed for Sling TV great work!
Congrats and keep it up.
Letha commented on 10-Nov-2018 07:14 AM
With thanks! This What Equipment is Needed for Sling TV an amazing web page!
Shavonne commented on 10-Nov-2018 07:21 AM
You've good thing right.

Feel free to visit my web-siteWhat Equipment is Needed for Sling TV
Arnulfo commented on 10-Nov-2018 08:04 AM
You've gotten possibly the best webpages.


Also visit my web blog:What Equipment is Needed for Sling TV
188bet commented on 10-Nov-2018 09:41 AM
Should your motive here is to find paintings available for sale Melbourne or paintings
available for sale Brisbane, unfortunately nevertheless, you can't notice here.

Waterslide paper is provided in clear or white however clear
is much more preferred, considering that any type of unprinted locations for the image remains clear.

Then it matters not if it is heads or tail, you can predict a final
results.
child escorts commented on 10-Nov-2018 10:20 AM
Great blog! Is your theme custom made or did you download it from somewhere?
A design like yours with a few simple tweeks would really make my blog stand out.
Please let me know where you got your design. Thanks
keo nha cai commented on 10-Nov-2018 12:26 PM
It is common to find the ornamental painting and
sculptures with shapes depicting an interesting combination of different components from the artist's religious,
physical and cultural background. A vector path, it doesn't matter what the twists and turns are,
could be more elastic and scalable. It is maybe one of the most worldwide of mediums,
in the its practice along with its range.
Mayola Tomasulo commented on 10-Nov-2018 03:42 PM
Thanks for sharing..

Post a Comment




Captcha Image

Trackback Link
http://datascientists.com/BlogRetrieve.aspx?BlogID=10882&PostID=711829&A=Trackback
Trackbacks
Post has no trackbacks.

Can we unlock the Deep Learning black box?

Troy Sadkowsky - Wednesday, June 28, 2017

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Times; color: #000000; -webkit-text-stroke: #000000} p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Times; color: #000000; -webkit-text-stroke: #000000; min-height: 14.0px} span.s1 {font-kerning: none} span.s2 {text-decoration: underline ; font-kerning: none; color: #0000ff; -webkit-text-stroke: 0px #0000ff}

Can we unlock the Deep Learning black box?

Artificial intelligence has experienced a boom in recent years due to the increasing automation and generation of big data. Due to deep learning and artificial neural networks, machines are acquiring new perceptual abilities, such as recognizing images, speech, or reading handwriting, paving the way for an infinite number of new applications in human lives. However, this leap forward in AI processing comes at a cost – the reasoning behind the choices that deep learning networks make has become inscrutable to the engineers that built them.

Progress in automated vision greatly benefits from deep learning because general visual processing is far too complex to code by hand. A program that can learn by observing human example or through complicated training sets, and can generate its own algorithms through an interconnected network of a dozen to several hundred layers of simulated neurons, positions us that much closer to a self-driving car, or other AI programs capable of sophisticated automated decision-making. But if we cannot understand a program’s individual decisions – why it chose to drive into a tree, why a patient’s medications should be changed, why one individual was hired for the job and another flagged as a terrorist – what ethical and moral boundaries are we crossing? And if an artificial intelligence program is not 100% accurate, as they rarely are, what are the dangers of putting ourselves at the mercy of a machine’s mistakes?

According to an article in MIT Technology Review (https://www.technologyreview.com/s/604087/the-dark-secret-at-the-heart-of-ai/), trust will be a key factor in future applications of deep learning. Once machines can explain their reasoning to humans, we will better be able to learn from and act on their insights. Research is underway to develop tools for machine learning programs to be able to explain themselves. Regina Barzilay, at MIT, is developing a system that can collaborate with doctors by extracting snippets of text that represent patterns it has discover, and Carlos Guestrin’s system, from the University of Washington, highlights significant keywords or parts of an image to support a particular decision.

However, interpretable deep learning is far off, and even then, explanations offered by machines will always be simplified to a degree, which will always make trusting machine learning programs controversial. As a consequence, the European Union might soon pass legislation to make explanations for automated decisions a fundamental legal right, required of the companies that implement AI machine programs.

Jeff Clune, an investigator at the University of Wyoming testing deep neural networks, offers that, just as with human intelligence and decision-making, it is perhaps “the nature of intelligence that only part of it is exposed to rational explanation. Some of it is just instinctual, or subconscious, or inscrutable”. As deep learning offers more appealing possibilities in medicine, technology and other industries, when and should, society take the deep learning leap of faith?

Comments
Quentin commented on 09-Nov-2018 05:21 AM
Wow, gorgeous portal. Thnx ...

Here is my website;FIFA55
free android scales commented on 09-Nov-2018 05:49 AM
I blog quite often and I really thank you for your information. The article
has truly peaked my interest. I'm going to take a note of your blog and keep checking for new
details about once per week. I opted in for your Feed as well.
wanelo.co commented on 09-Nov-2018 07:27 AM
Everything is very open with a really clear description of the challenges.
It was definitely informative. Your website is very useful.
Many thanks for sharing!
Felica commented on 09-Nov-2018 09:05 AM
I like this site - its so usefull and helpfull.


My web siteQuest Bars Cheap
188bet commented on 09-Nov-2018 10:25 AM
This work reveals some sort of poetic mood and everyone would easily be attracted by it.
If this is a matter of yours too, then you definitely should learn concerning the best techniques
to procure such things. The beginning of Leonardo's life was
dedicated to art and painting in particular.
agen bola commented on 09-Nov-2018 01:00 PM
Appreciate this post. Will try it out.
바카라사이트 commented on 09-Nov-2018 01:33 PM
Aw, this was an extremely good post. Taking the time and actual effort to
create a great article? but what can I say?
I put things off a lot and never seem to get nearly anything done. https://www.banker9.net
stays on topic and states valid points. Thank you.
188bet commented on 09-Nov-2018 02:21 PM
Should your motive the following is to learn paintings available for sale Melbourne or paintings available Brisbane, unfortunately however, you can't view it here.

After the Bourbon Restoration, because trial participant of Louis XVI, David
was deprived of his civil right and property, and
was made to leave his homeland to stay in Brussels where David also completed many works, and lastly died in the strange land.
Then it is not important whether it is heads or tail,
one can predict a final results.
Francesca commented on 09-Nov-2018 05:30 PM
Hi there, well put together internet site you have got here.


my page:Can anybody use PLR articles for writing an eBook?
Juliane commented on 09-Nov-2018 05:49 PM
I really like what you guys are usually up too. This kind of clever work and coverage!
Keep up the wonderful works guys I've included you guys to my personal blogroll.

Also visit my site;Personal Struggle
https://github.com/jasonhocky/vincity/issues/1 commented on 09-Nov-2018 06:37 PM
I'm now not sure where you're getting your information, however good topic.
I needs to spend a while finding out much more or figuring
out more. Thanks for excellent info I used to be searching for this info
for my mission.
Your site has proven useful to me.
Teddy commented on 09-Nov-2018 08:37 PM
This excellent website really has all of the information I wanted concerning this subject and didn�t know who to ask.



My webpage;Https://Kumpulanitu.com
shopping online commented on 09-Nov-2018 10:40 PM
Hi there, I found your website via Google even as searching for a similar matter, your
web site came up, it seems to be great. I've bookmarked it in my google bookmarks.

Hello there, simply was aware of your blog through Google,
and located that it is truly informative.
I'm gonna be careful for brussels. I'll be grateful when you continue this in future.
Lots of people will be benefited out of your writing.
Cheers!
keo nha cai commented on 10-Nov-2018 12:39 AM
Should your motive here is to find out paintings available Melbourne or
paintings for sale Brisbane, unfortunately nevertheless, you can't see it here.
Waterslide paper emerged in clear or white however clear is more
preferred, given that almost any unprinted locations around the
image remains clear. Then it does not matter if it is heads or tail, it's possible to predict the ultimate results.
Shayne commented on 10-Nov-2018 06:52 AM
Wow cuz this What Equipment is Needed for Sling TV great work!
Congrats and keep it up.
Letha commented on 10-Nov-2018 07:14 AM
With thanks! This What Equipment is Needed for Sling TV an amazing web page!
Shavonne commented on 10-Nov-2018 07:21 AM
You've good thing right.

Feel free to visit my web-siteWhat Equipment is Needed for Sling TV
Arnulfo commented on 10-Nov-2018 08:04 AM
You've gotten possibly the best webpages.


Also visit my web blog:What Equipment is Needed for Sling TV
188bet commented on 10-Nov-2018 09:41 AM
Should your motive here is to find paintings available for sale Melbourne or paintings
available for sale Brisbane, unfortunately nevertheless, you can't notice here.

Waterslide paper is provided in clear or white however clear
is much more preferred, considering that any type of unprinted locations for the image remains clear.

Then it matters not if it is heads or tail, you can predict a final
results.
child escorts commented on 10-Nov-2018 10:20 AM
Great blog! Is your theme custom made or did you download it from somewhere?
A design like yours with a few simple tweeks would really make my blog stand out.
Please let me know where you got your design. Thanks
keo nha cai commented on 10-Nov-2018 12:26 PM
It is common to find the ornamental painting and
sculptures with shapes depicting an interesting combination of different components from the artist's religious,
physical and cultural background. A vector path, it doesn't matter what the twists and turns are,
could be more elastic and scalable. It is maybe one of the most worldwide of mediums,
in the its practice along with its range.
Mayola Tomasulo commented on 10-Nov-2018 03:42 PM
Thanks for sharing..

Post a Comment




Captcha Image

Trackback Link
http://datascientists.com/BlogRetrieve.aspx?BlogID=10882&PostID=711829&A=Trackback
Trackbacks
Post has no trackbacks.