Easily Fooling Deep Neural Networks
Data Skeptic16 Jan 2015

Easily Fooling Deep Neural Networks

My guest this week is Anh Nguyen, a PhD student at the University of Wyoming working in the Evolving AI lab. The episode discusses the paper Deep Neural Networks are Easily Fooled [pdf] by Anh Nguyen, Jason Yosinski, and Jeff Clune. It describes a process for creating images that a trained deep neural network will mis-classify. If you have a deep neural network that has been trained to recognize certain types of objects in images, these "fooling" images can be constructed in a way which the network will mis-classify them. To a human observer, these fooling images often have no resemblance whatsoever to the assigned label. Previous work had shown that some images which appear to be unrecognizable white noise images to us can fool a deep neural network. This paper extends the result showing abstract images of shapes and colors, many of which have form (just not the one the network thinks) can also trick the network.

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(601)

Emoji as a Predictor

Emoji as a Predictor

Emojis are arguably one of the most effective ways to express emotions when texting. In today's episode, Xuan Lu shares her research on the use of emojis by developers. She explains how the study of e...

23 Mai 202221min

Polarizing Trends in the Gig Economy

Polarizing Trends in the Gig Economy

On the show today, Fabian Braesemann, a research fellow at the University of Oxford, joins us to discuss his study analyzing the gig economy. He revealed the trends he discovered since remote work bec...

16 Mai 202246min

Remote Learning in Applied Engineering

Remote Learning in Applied Engineering

On the show today, we interview Mouhamed Abdulla, a professor of Electrical Engineering at Sheridan Institute of Technology. Mouhamed joins us to discuss his study on remote teaching and learning in a...

12 Mai 202225min

Remote Productivity

Remote Productivity

It is difficult to estimate the effect on remote working across the board. Darja Šmite, who speaks with us today, is a professor of Software Engineering at the Blekinge Institute of Technology. In her...

9 Mai 202229min

Does Remote Learning Work?

Does Remote Learning Work?

We explore this complex question in two interviews today.  First, Kasey Wagoner describes 3 approaches to remote lab sessions and an analysis of which was the most instrumental to students.  Second, T...

1 Mai 202248min

Covid-19 Impact on Bicycle Usage

Covid-19 Impact on Bicycle Usage

In this episode, we speak with Abdullah Kurkcu, a Lead Traffic Modeler. Abdullah joins us to discuss his recent study on the effect of COVID-19 on bicycle usage in the US. He walks us through the data...

25 Apr 202231min

Learning Digital Fabrication Remotely

Learning Digital Fabrication Remotely

Today, we are joined by Jennifer Jacobs and Nadya Peek, who discuss their experience in teaching remote classes for a course that is largely hands-on. The discussion was focused on digital fabrication...

22 Apr 202233min

Remote Software Development

Remote Software Development

Today, we are joined by Denae Ford, a Senior Researcher at Microsoft Research and an Affiliate Assistant Professor at the University of Washington. Denae discusses her work around remote work and its ...

18 Apr 202237min

Populært innen Vitenskap

fastlegen
jss
tingenes-tilstand
forskningno
rekommandert
rss-paradigmepodden
villmarksliv
sinnsyn
rss-zahid-ali-hjelper-deg
vett-og-vitenskap-med-gaute-einevoll
kvinnehelsepodden
rss-inn-til-kjernen-med-sunniva-rose
nordnorsk-historie
nevropodden
tidlose-historier
liberal-halvtime
fjellsportpodden
rss-overskuddsliv
grunnstoffene
rss-rekommandert