{"id":1444,"date":"2018-04-03T15:11:38","date_gmt":"2018-04-03T15:11:38","guid":{"rendered":"https:\/\/blogs.mathworks.com\/headlines\/?p=1444"},"modified":"2021-09-22T15:22:32","modified_gmt":"2021-09-22T19:22:32","slug":"bug-brain-beats-machine-learning","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/headlines\/2018\/04\/03\/bug-brain-beats-machine-learning\/","title":{"rendered":"Bug Brain Beats Machine Learning"},"content":{"rendered":"

Living organisms have long provided inspiration for technology. Biomimicry of birds helped us design our first aircraft, while the structure of seed burs was copied for Velcro. Today, biomimicry is being applied to advanced technology such as robotics<\/a> and computer vision<\/a>.<\/p>\n

Another modern-day application of biomimicry is in artificial intelligence (AI).\u00a0 With AI, machines take on natural cognitive functions such as learning and problem-solving. Artificial neural networks (ANNs) take biomimicry a step further by creating computing systems inspired by the brains of living organisms.<\/p>\n

But just how intelligent can a system be that is modeled after a relatively unsophisticated biological brain? It turns out, thanks to evolution, even relatively simple brains of living creatures can be very intelligent when it comes to a task that is necessary for their survival. For a moth, that means the sense of smell.<\/p>\n

\"Close-up<\/a><\/h2>\n

<\/h2>\n

Sometimes, smaller is better<\/h2>\n

Even though a moth\u2019s brain is the size of a pinhead<\/a>, it is highly efficient when it comes to learning new odors. Its sense of smell is needed for finding food and finding mates, both critical tasks for their survival as a species.<\/p>\n

Researchers from the University of Washington developed a neural network, dubbed MothNet, based on the structure of a moth\u2019s brain.<\/p>\n

\u201cThe moth olfactory network is among the simplest biological neural systems that can learn,\u201d the researchers, Charles B. Delahunt, Jeffrey Riffell, and J. Nathan Kutz, stated in their paper, Biological Mechanisms for Learning: A Computational Model of Olfactory Learning in the Manduca sexta Moth, with Applications to Neural Nets.<\/a><\/p><\/blockquote>\n

The MIT Technology Review<\/em> article, Why even a moth\u2019s brain is smarter than AI<\/a>, described the biological system copied by MothNet, \u201cThe olfactory learning system in moths is relatively simple and well mapped by neuroscientists. It consists of five distinct networks that feed information forward from one to the next.\u201d<\/p>\n

 <\/p>\n

\"Diagram<\/a>

MothNet was designed to mimic the structure of a moth\u2019s brain. Image credit: Delahunt and Kutz.<\/p><\/div><\/p>\n

 <\/p>\n

Instead of identifying scents, the researchers used supervised learning to train the ANN to recognize handwritten numbers with just 15 to 20 images of each digit from zero to nine. The training samples came from the MNIST (Modified National Institute of Standards and Technology) database of digits commonly used for training and testing in the field of machine learning. Some Examples from the MNIST database are shown below:<\/p>\n

 <\/p>\n

\"Handwritten<\/a>

Sample from the MINST database. Image credit: Wikipedia, CC 4.0.<\/p><\/div><\/p>\n

They found MothNet could learn much faster than machines.\u00a0 MothNet \u201clearned\u201d to recognize numbers with just a few training samples with an accuracy of 75 to 85 percent. A typical convolutional neural network, by comparison, requires thousands of training examples to achieve 99% accuracy.<\/p>\n

Developing Better Machine Learning Algorithms<\/h2>\n

The researchers found that the moth\u2019s biological system was efficient at learning due to three main characteristics which could aid in the development of new machine learning algorithms:<\/p>\n