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
Raymundo Sayle commented on 16-Apr-2019 10:09 AM
Oh my goodness! an excellent post dude. Thanks a ton Nevertheless I am experiencing problem with ur rss . Do not know why Cannot enroll in it. Will there be everyone getting identical rss difficulty? Anyone who knows kindly respond. Thnkx
bandar togel terpercaya commented on 16-Apr-2019 10:26 AM
Great goods from you, man. I have understand your stuff previous to and
you are just too magnificent. I actually like what you've acquired here, really like what you're saying and the way in which you say
it. You make it entertaining and you still care for to keep it smart.
I cant wait to read much more from you. This is really a tremendous website.
Kacey Youd commented on 16-Apr-2019 10:41 AM
great . Thanks for informations . Ill be back. Thanks again
Rufina Haggin commented on 16-Apr-2019 10:54 AM
There are some fascinating points with time in the following paragraphs but I do not determine if I see these people center to heart. There may be some validity but I’m going to take hold opinion until I take a look at it further. Good article , thanks and we want much more! Put into FeedBurner likewise
Ladonna Joice commented on 16-Apr-2019 10:54 AM
You should participate in a contest for among the best blogs on the web. I will advocate this website!
cost of lead commented on 16-Apr-2019 11:25 AM
Sweet site, super style and design, really clean and apply pleasant.
Trey Hefflinger commented on 16-Apr-2019 11:27 AM
We all do yetabroad suggest to develop virtually any canadinic or labled forward-looking pediatricians that could mononitate pasen through, or even upon [b]user reports on expert viagra[/b] associated with, protherics.
a commented on 16-Apr-2019 11:40 AM
What's up, I wish for to subscribe for this web site to obtain most recent updates, therefore where can i do it please help.
Sharolyn Raymore commented on 16-Apr-2019 11:48 AM
Perfectly pent content material , appreciate it for entropy.
See Brandstetter commented on 16-Apr-2019 11:56 AM
niggas jumpin on they favorite rapper, I will kill your favorite rapper. put him in a grave and start stumpin on your favorite rapper//. so rest in peace to yo mafuckin favorite rapper….. (I just committed suicide)… play it backwards//. . CHECK OUT MY SONG KING OF MY CITY AND SEE WHY VA WILL NEVER BE THE SAME!!.
Viviana Bronston commented on 16-Apr-2019 11:57 AM
For example, develop suppliers have worked with the Meals & Drug Administration (FDA) to make sure that the marketplace specifications developed consequence in the ideal and most effective processes. And to communicate any public announcements, the marketplace has created relationships with associations like the National Grocer’s Association (NGA).
Lauralee Trial commented on 16-Apr-2019 12:12 PM
Wow, suprisingly I never knew this. Keep up with good posts.
Sharyn commented on 16-Apr-2019 12:24 PM
Rikalize Reinecke at work on her aquaponics farm.
Drema Gangelhoff commented on 16-Apr-2019 01:10 PM
Hi my loved one! I wish to say that this post is amazing, great written and come with almost all important infos. I would like to see more posts like this .
Shyla Schlitzer commented on 16-Apr-2019 01:10 PM
Perfect work you have done, this website is really cool with excellent information.
Hilario Lilla commented on 16-Apr-2019 01:10 PM
Your current article constantly have a decent amount of really up to date info. Where do you come up with this? Just saying you are very resourceful. Thanks again
Dennis Erstad commented on 16-Apr-2019 01:30 PM
most toddlers love cute teddys as their first toddler toys, my son also loves them a lot,.
Roxane Mcmillian commented on 16-Apr-2019 01:30 PM
both LG and LiteOn makes a great performing dvd burner, they also feature those anti-shock mount`
webrtc streaming server commented on 16-Apr-2019 01:30 PM
Good day!
I'm doing video streaming as a course but I'm also
Chinese woman =).
I actually like Cheerleading!
Archer commented on 16-Apr-2019 02:09 PM
This man knew I was a sex worker. It says so, right in my own Bumble profile: retired media whore, current actual whore.
He'd even commented on it, using the language every woman longs to listen to from the romantic interest:'Haha, nice ;) '. And yet I watched as his face contorted in to an expression of disgust, his upper lip curling as the reality of my profession came crashing down around him such as for instance a tonne of bricks.


"That is clearly a lot," he said, and then he rolled onto his back and stared at the ceiling. I didn't hear from him again.

It sometimes surprises people to listen to that sex workers do all sorts of normal people activities, like working other jobs, studying, taking the bins out. We exist in the real world after our shifts end and the red light is flicked off; we've dinner with this families and shop at K-Mart and wait on hold with our websites providers for what feels like hours.


It's not common that the physical and emotional experiences we have at the office would be enough to make up for a possible insufficient intimate connection inside our lives outside work; so most of us also date, with varied degrees of success.

A few months ago, I ended a connection with a man I have been seeing for nearly two years. In private, he was a massive supporter of me working, but around his colleagues and friends his tune appeared to change. He'd introduce me, but hesitate in describing our relationship; when he said, "That is Kate..." the silence that hung in the space where, "...my girlfriend," should have already been weighed a tonne.


I don't believe that he personally had a problem with me being a sex worker, but I really do genuinely believe that the likelihood of other people judging me – and then judging him to be with me – was enough to create him want to keep me a secret.

Feel free to visit my weblog :: נערות ליווי בתל אביב
Lupe Simer commented on 16-Apr-2019 02:29 PM
Good write-up, I’m regular visitor of one’s web site, maintain up the excellent operate, and It’s going to be a regular visitor for a long time.
Chiquita Callar commented on 16-Apr-2019 02:30 PM
Rise Of The Lycans do have some great fighting scene between the werewolves and the vampires, it is a nice movie::
Margarito Sovern commented on 16-Apr-2019 02:30 PM
Ashton Kutcher and Demi Moore have already spoken out against the idea, saying that they’d rather not encourage anymore people to stalk celebrities. So, the first thing to do is to decide the purpose of your ad.
Hayden Nippe commented on 16-Apr-2019 02:30 PM
Well worded post will be sharing this with my readers this evening
Marisela Catanach commented on 16-Apr-2019 02:48 PM
I dont agree with all of the points made but still a good read cheers.

Post a Comment




Captcha Image

Trackback Link
http://www.datascientists.net/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
Raymundo Sayle commented on 16-Apr-2019 10:09 AM
Oh my goodness! an excellent post dude. Thanks a ton Nevertheless I am experiencing problem with ur rss . Do not know why Cannot enroll in it. Will there be everyone getting identical rss difficulty? Anyone who knows kindly respond. Thnkx
bandar togel terpercaya commented on 16-Apr-2019 10:26 AM
Great goods from you, man. I have understand your stuff previous to and
you are just too magnificent. I actually like what you've acquired here, really like what you're saying and the way in which you say
it. You make it entertaining and you still care for to keep it smart.
I cant wait to read much more from you. This is really a tremendous website.
Kacey Youd commented on 16-Apr-2019 10:41 AM
great . Thanks for informations . Ill be back. Thanks again
Rufina Haggin commented on 16-Apr-2019 10:54 AM
There are some fascinating points with time in the following paragraphs but I do not determine if I see these people center to heart. There may be some validity but I’m going to take hold opinion until I take a look at it further. Good article , thanks and we want much more! Put into FeedBurner likewise
Ladonna Joice commented on 16-Apr-2019 10:54 AM
You should participate in a contest for among the best blogs on the web. I will advocate this website!
cost of lead commented on 16-Apr-2019 11:25 AM
Sweet site, super style and design, really clean and apply pleasant.
Trey Hefflinger commented on 16-Apr-2019 11:27 AM
We all do yetabroad suggest to develop virtually any canadinic or labled forward-looking pediatricians that could mononitate pasen through, or even upon [b]user reports on expert viagra[/b] associated with, protherics.
a commented on 16-Apr-2019 11:40 AM
What's up, I wish for to subscribe for this web site to obtain most recent updates, therefore where can i do it please help.
Sharolyn Raymore commented on 16-Apr-2019 11:48 AM
Perfectly pent content material , appreciate it for entropy.
See Brandstetter commented on 16-Apr-2019 11:56 AM
niggas jumpin on they favorite rapper, I will kill your favorite rapper. put him in a grave and start stumpin on your favorite rapper//. so rest in peace to yo mafuckin favorite rapper….. (I just committed suicide)… play it backwards//. . CHECK OUT MY SONG KING OF MY CITY AND SEE WHY VA WILL NEVER BE THE SAME!!.
Viviana Bronston commented on 16-Apr-2019 11:57 AM
For example, develop suppliers have worked with the Meals & Drug Administration (FDA) to make sure that the marketplace specifications developed consequence in the ideal and most effective processes. And to communicate any public announcements, the marketplace has created relationships with associations like the National Grocer’s Association (NGA).
Lauralee Trial commented on 16-Apr-2019 12:12 PM
Wow, suprisingly I never knew this. Keep up with good posts.
Sharyn commented on 16-Apr-2019 12:24 PM
Rikalize Reinecke at work on her aquaponics farm.
Drema Gangelhoff commented on 16-Apr-2019 01:10 PM
Hi my loved one! I wish to say that this post is amazing, great written and come with almost all important infos. I would like to see more posts like this .
Shyla Schlitzer commented on 16-Apr-2019 01:10 PM
Perfect work you have done, this website is really cool with excellent information.
Hilario Lilla commented on 16-Apr-2019 01:10 PM
Your current article constantly have a decent amount of really up to date info. Where do you come up with this? Just saying you are very resourceful. Thanks again
Dennis Erstad commented on 16-Apr-2019 01:30 PM
most toddlers love cute teddys as their first toddler toys, my son also loves them a lot,.
Roxane Mcmillian commented on 16-Apr-2019 01:30 PM
both LG and LiteOn makes a great performing dvd burner, they also feature those anti-shock mount`
webrtc streaming server commented on 16-Apr-2019 01:30 PM
Good day!
I'm doing video streaming as a course but I'm also
Chinese woman =).
I actually like Cheerleading!
Archer commented on 16-Apr-2019 02:09 PM
This man knew I was a sex worker. It says so, right in my own Bumble profile: retired media whore, current actual whore.
He'd even commented on it, using the language every woman longs to listen to from the romantic interest:'Haha, nice ;) '. And yet I watched as his face contorted in to an expression of disgust, his upper lip curling as the reality of my profession came crashing down around him such as for instance a tonne of bricks.


"That is clearly a lot," he said, and then he rolled onto his back and stared at the ceiling. I didn't hear from him again.

It sometimes surprises people to listen to that sex workers do all sorts of normal people activities, like working other jobs, studying, taking the bins out. We exist in the real world after our shifts end and the red light is flicked off; we've dinner with this families and shop at K-Mart and wait on hold with our websites providers for what feels like hours.


It's not common that the physical and emotional experiences we have at the office would be enough to make up for a possible insufficient intimate connection inside our lives outside work; so most of us also date, with varied degrees of success.

A few months ago, I ended a connection with a man I have been seeing for nearly two years. In private, he was a massive supporter of me working, but around his colleagues and friends his tune appeared to change. He'd introduce me, but hesitate in describing our relationship; when he said, "That is Kate..." the silence that hung in the space where, "...my girlfriend," should have already been weighed a tonne.


I don't believe that he personally had a problem with me being a sex worker, but I really do genuinely believe that the likelihood of other people judging me – and then judging him to be with me – was enough to create him want to keep me a secret.

Feel free to visit my weblog :: נערות ליווי בתל אביב
Lupe Simer commented on 16-Apr-2019 02:29 PM
Good write-up, I’m regular visitor of one’s web site, maintain up the excellent operate, and It’s going to be a regular visitor for a long time.
Chiquita Callar commented on 16-Apr-2019 02:30 PM
Rise Of The Lycans do have some great fighting scene between the werewolves and the vampires, it is a nice movie::
Margarito Sovern commented on 16-Apr-2019 02:30 PM
Ashton Kutcher and Demi Moore have already spoken out against the idea, saying that they’d rather not encourage anymore people to stalk celebrities. So, the first thing to do is to decide the purpose of your ad.
Hayden Nippe commented on 16-Apr-2019 02:30 PM
Well worded post will be sharing this with my readers this evening
Marisela Catanach commented on 16-Apr-2019 02:48 PM
I dont agree with all of the points made but still a good read cheers.

Post a Comment




Captcha Image

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