MAY 27, 2021 9:08 AM PDT

Just add noise: improving AI decision-making

New work on artificial intelligence from a collaboration between the University of Texas at San Antonio (UTSA), the University of Central Florida (UCF), the Air Force Research Laboratory (AFRL) and SRI International improves the way that AI learns. The method alters the way that machine learning decisions are made by adding noise (also called pixilation) into multiple layers of a neural network. 

"It's about injecting noise into every layer," said lead researcher Sumit Jha, a professor in the Department of Computer Science at UTSA. "The network is now forced to learn a more robust representation of the input in all of its internal layers. If every layer experiences more perturbations in every training, then the image representation will be more robust and you won't see the AI fail just because you change a few pixels of the input image."

This work builds on Jha’s team’s past investigations into AI safety, research which they presented in 2019 at the AI Safety workshop and the International Joint Conference on Artificial Intelligence (IJCAI). They say that developing "explainable AI,” which refers to artificial intelligence that has a high level of assurance, is imperative to trust this form of technology for applications such as medical imaging and autonomous driving.

The development from this team involves the use of stochastic differential equations (SDEs) overneural ordinary differential equations (ODEs) within a network. While ODEs train a machine with one input through one network which then moves throughout layers to create one response in the output layer, SEDs learn from a set of images as a result of the injection of noise in multiple layers of the neural network.

The team plans to present this advancement, which they describe in the paper "On Smoother Attributions using Neural Stochastic Differential Equations," at the next IJCAI. 

"I am delighted to share the fantastic news that our paper on explainable AI has just been accepted at IJCAI," Jha exclaims. "This is a big opportunity for UTSA to be part of the global conversation on how a machine sees."

Sources: USTA, Eureka Alert

About the Author
  • Kathryn is a curious world-traveller interested in the intersection between nature, culture, history, and people. She has worked for environmental education non-profits and is a Spanish/English interpreter.
You May Also Like
MAR 03, 2021
Chemistry & Physics
Exploring the swirlonic state of active matter
MAR 03, 2021
Exploring the swirlonic state of active matter
Research published in the journal Scientific Reports describes a new kind of active matter, termed a swirlonic state. As ...
MAR 10, 2021
Chemistry & Physics
Where does the energy go in singlet fission?
MAR 10, 2021
Where does the energy go in singlet fission?
New research from scientists at Linköping University, Sweden, describes a recent discovery revealing where the ener ...
APR 19, 2021
Chemistry & Physics
Breaking down the mathematical laws of forests
APR 19, 2021
Breaking down the mathematical laws of forests
Researchers from the Santa Fe Institute (SFI) have published a new study in the Proceedings of the National Academy of S ...
MAY 22, 2021
Plants & Animals
This Lizard Blows Bubbles to Breathe Underwater
MAY 22, 2021
This Lizard Blows Bubbles to Breathe Underwater
Scientists say it's the biological version of the rebreathers than human divers use to conserve and recycle their oxygen ...
MAY 29, 2021
Space & Astronomy
First Matter in the Universe Flowed Like Tap Water
MAY 29, 2021
First Matter in the Universe Flowed Like Tap Water
In two separate studies, researchers led by those at the University of Copenhagen and Queen Mary University of London fo ...
JUN 11, 2021
Chemistry & Physics
Cutting-edge wearables: the next generation of electronics
JUN 11, 2021
Cutting-edge wearables: the next generation of electronics
New research published in Applied Physics Reviews from AIP Publishing considers the development of flexible supercapacit ...
Loading Comments...