A historical survey of algorithms and hardware architectures for neural-inspired and neuromorphic computing applications

Abstract

Biological neural networks continue to inspire new developments in algorithms and microelectronic hardware to solve challenging data processing and classification problems. Here, we survey the history of neural-inspired and neuromorphic computing in order to examine the complex and intertwined trajectories of the mathematical theory and hardware developed in this field. Early research focused on adapting existing hardware to emulate the pattern recognition capabilities of living organisms. Contributions from psychologists, mathematicians, engineers, neuroscientists, and others were crucial to maturing the field from narrowly-tailored demonstrations to more generalizable systems capable of addressing difficult problem classes such as object detection and speech recognition. Algorithms that leverage fundamental principles found in neuroscience such as hierarchical structure, temporal integration, and robustness to error have been developed, and some of these approaches are achieving world-leading performance on particular data classification tasks. In addition, novel microelectronic hardware is being developed to perform logic and to serve as memory in neuromorphic computing systems with optimized system integration and improved energy efficiency. Key to such advancements was the incorporation of new discoveries in neuroscience research, the transition away from strict structural replication and towards the

The Power of Computation-in-Memory Based on Memristive Devices

“…Many alternatives architectures and technologies are under investigations. Resistive computing [4], quantum computing [5], and neuromorphic computing [6] are couple of alternative computing notions, while memristive devices, quantum dots, spin-wave devices are couple of emerging device technologies [7]. Memristive device is seen as a very a promising candidate to complement and/or replace traditional CMOS (at least in some applications) due to many advantages including CMOS process compatibility [8], zero standby power, great scalability, and high density, etc.…”

Unsupervised visual feature learning with spike-timing-dependent plasticity: How far are we from traditional feature learning approaches?

“…Initially studied in neuroscience as a model of the brain, SNNs receive constant attention in the fields of machine learning and pattern recognition, from both the theoretical [22] and the applicative [17; 23; 24; 25] perspectives. Dedicated hardware implementing this model can be very energy-efficient [20]. SNNs have already shown their ability to provide near-state-of-the-art results in image classification, but only when they are trained by transferring parameters from pre-trained deep neural networks [21] or by variants of back-propagation [26].…”

The non-capacitor model of leaky integrate-and-fire VO2 neuron with the thermal mechanism of the membrane potential

“…In recent decades, fundamental and applied research in the field of artificial intelligence has been actively conducted, where artificial neural networks (ANNs) play the leading role [1,2]. The main efforts are focused at developing new architectures, optimal learning algorithms and ways to improve the accuracy of ANN operation [3,4].…”