Table 2.
Key glossary
| Term | Abbreviation | Definition |
|---|---|---|
| Self-rectifying memristor | SRM | A novel type of memristor that exhibits intrinsic diode-like rectification, enabling unidirectional conduction and suppressing sneak path currents in crossbar arrays |
| Rectification ratio | RR | The ratio of the on-state current of the device under positive bias to the off-state current under negative bias, indicating the level of current suppression in reverse bias |
| Nonlinearity | NL | The ratio of the current of the device at the read voltage under the low-resistance state (LRS) to its current at the partial read voltage, indicating the degree of nonlinearity in the current–voltage characteristics |
| Complementary metal–oxide–semiconductor | CMOS | A widely used technology for manufacturing integrated circuits, characterized by low-power consumption and high scalability |
| Vector–matrix multiplication | VMM | A fundamental operation in many computing tasks, where a vector is multiplied by a matrix to produce a result vector, often used in neural networks and in-memory computing |
| Ternary content-addressable memory | TCAM | A type of cell that allows data to be retrieved based on its content rather than its address, often used in high-speed search applications |
| Artificial neural network | ANN | A computational model inspired by the structure and function of biological neural networks, used for tasks such as image recognition and pattern classification |
| Convolutional neural network | CNN | A type of neural network that uses convolutional layers to process data with grid-like topology, commonly used for image and video recognition tasks |
| Autonomous driving systems | ADS | Systems that enable vehicles to operate without human intervention, often using advanced sensors, computing, and machine learning techniques |
| Reservoir computing | RC | A type of recurrent neural network that uses a fixed, highly dynamic reservoir to map input signals to a high-dimensional space, followed by a linear readout layer for output |
| Physical unclonable function | PUF | A security primitive that generates unique and unclonable digital fingerprints based on the inherent physical variations in a device, used for authentication and key storage |
| True random number generator | TRNG | A hardware device that generates random numbers based on physical processes, providing high entropy for cryptographic applications |
| Homomorphic encryption | HE | A form of encryption that allows computations to be performed on ciphertext, producing an encrypted result that, when decrypted, matches the result of operations performed on the plaintext |