Artificial brains learn to adapt

May 16, 2014, National Science Foundation
Silvia Ferrari and her team at Duke University are creating new control and navigation systems for aircraft, robots and other engineered systems that emulate the brain's ability to adapt movement based on changing environmental conditions. They are designing artificial brains, known as spiking neural networks, that incorporate characteristic details of biological brains. Hippocampal cells (pictured) are especially important for navigation. Credit: Silvia Ferrari, Mechanical Engineering and Materials Science, Duke University

For every thought or behavior, the brain erupts in a riot of activity, as thousands of cells communicate via electrical and chemical signals. Each nerve cell influences others within an intricate, interconnected neural network. And connections between brain cells change over time in response to our environment.

Despite supercomputer advances, the remains the most flexible, efficient information processing device in the world. Its exceptional performance inspires researchers to study and imitate it as an ideal of computing power.

Artificial neural networks

Computer models built to replicate how the brain processes, memorizes and/or retrieves information are called artificial neural networks. For decades, engineers and computer scientists have used artificial neural networks as an effective tool in many real-world problems involving tasks such as classification, estimation and control.

However, artificial neural networks do not take into consideration some of the basic characteristics of the human brain such as signal transmission delays between neurons, membrane potentials and synaptic currents.

A new generation of neural network models—called spiking neural networks—are designed to better model the dynamics of the brain, where neurons initiate signals to other neurons in their networks with a rapid spike in cell voltage. In modeling biological neurons, spiking neural networks may have the potential to mimick brain activities in simulations, enabling researchers to investigate neural networks in a biological context.

With funding from the National Science Foundation, Silvia Ferrari of the Laboratory for Intelligent Systems and Controls at Duke University uses a new variation of spiking neural networks to better replicate the behavioral learning processes of mammalian brains.

Behavioral learning involves the use of sensory feedback, such as vision, touch and sound, to improve motor performance and enable people to respond and quickly adapt to their changing environment.

Silvia Ferrari and her team at Duke University trained a virtual insect whose nervous system is modeled by a large spiking neural network. The virtual insect was trained with an algorithm that responds to sensory feedback, such as vision, touch and sound. The virtual insect was able to adapt to changing conditions as it navigated the environment for virtual food. Credit: Gary W. Meek, Georgia Tech

"Although existing engineering systems are very effective at controlling dynamics, they are not yet capable of handling unpredicted damages and failures handled by biological brains," Ferrari said.

How to teach an artificial brain

Ferrari's team is applying the spiking neural network model of learning on the fly to complex, critical , such as aircraft and power plants, with the goal of making them safer, more cost-efficient and easier to operate.

The team has constructed an algorithm that teaches spiking neural networks which information is relevant and how important each factor is to the overall goal. Using computer simulations, they've demonstrated the algorithm on aircraft flight control and robot navigation.

They started, however, with an insect.

"Our method has been tested by training a virtual insect to navigate in an unknown terrain and find foods," said Xu Zhang, a Ph.D. candidate who works on training the spiking neural network. "The nervous system was modeled by a large spiking neural network with unknown and random synaptic connections among those neurons."

Having tested their algorithm in computer simulations, they now are in the process of testing it biologically.

To do so, they will use lab-grown brain cells genetically altered to respond to certain types of light. This technique, called optogenetics, allows researchers to control how nerve cells communicate. When the light pattern changes, the neural activity changes.

Graduate student Xu Zhang, postdoc Greg Foderaro and Professor Silvia Ferrari (pictured left to right) at Duke University are translating the abilities of biological brains into computational capabilities to solve complex control and estimation problems. Credit: Silvia Ferrari, Mechanical Engineering and Materials Science, Duke University

The researchers hope to observe that the living adapts over time to the light patterns and therefore have the ability to store and retrieve sensory information, just as human neuronal networks do.

Large-scale applications of small-scale findings

Uncovering the fundamental mechanisms responsible for the brain's learning processes can potentially yield insights into how humans learn—and make an everyday difference in people's lives.

Such insights may advance the development of certain artificial devices that can substitute for certain motor, sensory or cognitive abilities, particularly prosthetics that respond to feedback from the user and the environment. People with Parkinson's disease and epilepsy have already benefited from these types of devices.

"One of the most significant challenges in reverse-engineering the brain is to close the knowledge gap that exists between our understanding of biophysical models of neuron-level activity and the synaptic plasticity mechanisms that drive meaningful learning," said Greg Foderaro, a postdoctoral fellow involved the the research.

"We believe that by considering the networks at several levels—from computation to cell cultures to brains—we can greatly expand our understanding of the system of sensory and motor functions, as well as making a large step towards understanding the as a whole."

Explore further: Chips that mimic the brain

Related Stories

Chips that mimic the brain

July 22, 2013
No computer works as efficiently as the human brain – so much so that building an artificial brain is the goal of many scientists. Neuroinformatics researchers from the University of Zurich and ETH Zurich have now made ...

Researchers connect neurons to computers to decipher the enigmatic code of neuronal circuits

July 12, 2011
Machine logic is based on human logic. But although a computer processor can be dissembled and dissected in logical steps, the same is not true for the way our brains process information, says Mark Shein of Tel Aviv University's ...

K computer is unleashed for largest neuronal network simulation ever

January 14, 2014
(Medical Xpress)—Wow, today's top supercomputers have formidable speed and computational power to rival humans. On the other hand, wow, the human brain is more powerful than today's supercomputers, which with all their ...

Delving deep into the brain

May 1, 2014
Launched in 2013, the national BRAIN Initiative aims to revolutionize our understanding of cognition by mapping the activity of every neuron in the human brain, revealing how brain circuits interact to create memories, learn ...

Researchers create a reference atlas for neural circuits in fruit fly larvae

March 28, 2014
(Medical Xpress)—A team of researchers from Johns Hopkins University and Janelia Farm Research Campus has developed a new technique for studying neural circuits that helps tie circuit activity to organism behavior. In their ...

Recommended for you

Cognitive training helps regain a younger-working brain

January 23, 2018
Relentless cognitive decline as we age is worrisome, and it is widely thought to be an unavoidable negative aspect of normal aging. Researchers at the Center for BrainHealth at The University of Texas at Dallas, however, ...

Lifting the veil on 'valence,' brain study reveals roots of desire, dislike

January 23, 2018
The amygdala is a tiny hub of emotions where in 2016 a team led by MIT neuroscientist Kay Tye found specific populations of neurons that assign good or bad feelings, or "valence," to experience. Learning to associate pleasure ...

Your brain responses to music reveal if you're a musician or not

January 23, 2018
How your brain responds to music listening can reveal whether you have received musical training, according to new Nordic research conducted in Finland (University of Jyväskylä and AMI Center) and Denmark (Aarhus University).

New neuron-like cells allow investigation into synthesis of vital cellular components

January 22, 2018
Neuron-like cells created from a readily available cell line have allowed researchers to investigate how the human brain makes a metabolic building block essential for the survival of all living organisms. A team led by researchers ...

Finding unravels nature of cognitive inflexibility in fragile X syndrome

January 22, 2018
Mice with the genetic defect that causes fragile X syndrome (FXS) learn and remember normally, but show an inability to learn new information that contradicts what they initially learned, shows a new study by a team of neuroscientists. ...

Epilepsy linked to brain volume and thickness differences

January 22, 2018
Epilepsy is associated with thickness and volume differences in the grey matter of several brain regions, according to new research led by UCL and the Keck School of Medicine of USC.


Adjust slider to filter visible comments by rank

Display comments: newest first

3 / 5 (2) May 16, 2014
Artificial intelligence in the making.
1 / 5 (2) May 16, 2014
This needs to be shared more...especially the picture of the three researchers. It's important to see that not all Computer Scientists and Scientists and the like have to be geeky, ugly, or awkward..or male.
not rated yet May 18, 2014
The researchers hope to observe that the living neural network adapts over time to the light patterns and therefore have the ability to store and retrieve sensory information, just as human neuronal networks do.

If the damage and repair done (which the researchers here label adaptation in their model):

gives rise to expressions that no longer change over time they will have evidence of storage - or if you will - memory.

You need to represent DNA methylation as well otherwise all that is retrieved is volatile memory or volatile storage.

The researchers take an approach they believe will replicate their mentor - the neuron.
Be one of first to concede their success if their approach is right.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.