AI Definitions: Neural Networks
/Neural Networks (or artificial neural networks, ANNs) – A computer system modeled loosely on the human brain, inspired by brain cells (neurons). The artificial network finds complex associations, identifying patterns in text, images and sounds. In this type of machine learning, which underpins deep learning, scientists can train systems to “learn” human tasks by analyzing training examples. It processes information in layers, with the deepest layers doing the most complex work. Most recent developments in artificial intelligence have centered around neural networks (such as voice recognition, autonomous navigation, and drug discovery), though symbolic artificial intelligence was the dominant area of research for most of AI’s history.
Critics say our physical embodiment may be difficult if not impossible for symbolic processing systems to capture. If understanding is inseparable from physical experience, then the usefulness of neural networks will hit a limit.
Neural networks were first proposed in 1944 by two University of Chicago researchers (Warren McCullough and Walter Pitts). They moved to MIT in 1952 as founding members of what has been referred to as the first cognitive science department. Neural nets remained a major research area of neuroscience and computer science until 1969. A decade later, it enjoyed a resurgence, fell out of favor again in the first decade of the new century, and has since returned in the second decade, fueled by the increased processing power of graphics chips (or GPUs— graphical processing hardware) because of their use in video games.