Breaking Deep Learning Systems With Adversarial Examples | Two Minute Papers #43

Channel:
Hungary Two Minute Papers
Subscribers:
1,550,000
Published on ● Video Link: https://www.youtube.com/watch?v=j9FLOinaG94


Duration: 3:25
14,860 views
320

Artificial neural networks are computer programs that try to approximate what the human brain does to solve problems like recognizing objects in images. In this piece of work, the authors analyze the properties of these neural networks and try to unveil what exactly makes them think that a paper towel is a paper towel, and, building on this knowledge, try to fool these programs. Carefully crafted adversarial examples can be used to fool deep neural network reliably.

_______________

The paper "Intriguing properties of neural networks" is available here:
http://arxiv.org/abs/1312.6199

The paper "Explaining and Harnessing Adversarial Examples" is available here:
http://arxiv.org/abs/1412.6572

Image credits:
Thumbnail image - https://www.flickr.com/photos/healthblog/8384110298 (CC BY-SA 2.0)
Shower cap - Code Words / Julia Evans - https://codewords.recurse.com/issues/five/why-do-neural-networks-think-a-panda-is-a-vulture
MNIST - hxhl95

Andrej Karpathy's online convolutional neural network:
http://cs.stanford.edu/people/karpathy/convnetjs/demo/cifar10.html

Subscribe if you would like to see more of these! - http://www.youtube.com/subscription_center?add_user=keeroyz

Splash screen/thumbnail design: Felícia Fehér - http://felicia.hu

Károly Zsolnai-Fehér's links:
Patreon → https://www.patreon.com/TwoMinutePapers
Facebook → https://www.facebook.com/TwoMinutePapers/
Twitter → https://twitter.com/karoly_zsolnai
Web → https://cg.tuwien.ac.at/~zsolnai/



Other Videos By Two Minute Papers


2016-03-065000 Fellow Scholars Special! | Two Minute Papers
2016-03-02Awesome Research For Everyone! - Two Minute Papers Channel Trailer
2016-03-02How To Get Started With Machine Learning? | Two Minute Papers #51
2016-02-28Interactive Photo Recoloring | Two Minute Papers #50
2016-02-24Deep Learning Program Learns to Paint | Two Minute Papers #49
2016-02-21Artistic Manipulation of Caustics | Two Minute Papers #48
2016-02-18Should You Take the Stairs at Work? (For Weight Loss) | Two Minute Papers #47
2016-02-15What is Impostor Syndrome? | Two Minute Papers #46
2016-02-10Biophysical Skin Aging Simulations | Two Minute Papers #45
2016-02-06Extrapolations and Crowdfunded Research (Experiment) | Two Minute Papers #44
2016-02-03Breaking Deep Learning Systems With Adversarial Examples | Two Minute Papers #43
2016-01-31How DeepMind Conquered Go With Deep Learning (AlphaGo) | Two Minute Papers #42
2016-01-27What Do Virtual Objects Sound Like? | Two Minute Papers #41
2016-01-24Simulating Viscosity and Melting Fluids | Two Minute Papers #40
2016-01-20Interactive Editing of Subsurface Scattering | Two Minute Papers #39
2016-01-173D Printing Objects With Caustics | Two Minute Papers #38
2016-01-12Designing 3D Printable Robotic Creatures | Two Minute Papers #37
2016-01-09Designing Cities and Furnitures With Machine Learning | Two Minute Papers #36
2016-01-059 Cool Deep Learning Applications | Two Minute Papers #35
2015-12-31Neural Programmer-Interpreters Learn To Write Programs | Two Minute Papers #34
2015-12-24Peer Review #1 [Audio only] | Two Minute Papers


Tags:
two minute papers
Intriguing properties of neural networks
deep learning
deep neural network
generative adversarial networks
adversarial network
adversarial deep learning
adversarial neural network
neural network
artificial neural network
adversarial examples
breaking deep learning
Explaining and Harnessing Adversarial Examples
generative adverasial networks
neural network adversarial
generative adversarial nets
fooling deep learning
fooling neural network