Exploiting Ferroelectric and Spintronic Dynamics for Neural Network Computation
Ferroelectric and spintronic devices, relying on the control of polarization and magnetization, offer intrinsically fast, durable, energy‐efficient, and low‐latency building blocks for analog in‐memory computing. The hysteretic dynamics of an order parameter are leveraged to provide nonvolatile, multistate memory and nonlinear switching. Brain‐inspired
Dashiell Harrison +4 more
wiley +1 more source
Artificial Intelligence for Advanced Functional Materials: Progress and Emerging Frontiers
Artificial intelligence is transforming the discovery of functional materials by linking synthesis, characterization, simulation, and design in unified workflows. Advances in machine learning, autonomous experimentation, and foundation models are accelerating innovation across energy, electronics, and biomedicine, while revealing new frontiers for ...
Cristiano Malica +38 more
wiley +1 more source
Metal Oxide Nano-Interface Boosting the Deep Ultraviolet Adjustable Noise-Filtering In-Sensor Computing. [PDF]
Ju Z +9 more
europepmc +1 more source
Single‐Step Analog In‐Memory Matrix Computation in Three‐Dimensional Circuits
A wooden model illustrates a 3D memristive array designed for single‐step kernel convolutions. It shows how two‐dimensional input data are mapped into the 3D array and convolved in parallel with stored kernels. Color‐coded connection points represent programmed memory‐element values, with each color corresponding to a specific kernel weight.
Alireza Jaberi Rad +8 more
wiley +1 more source
Neuromorphic vision with quasi-BICs. [PDF]
Li J, Qin H, Song Q.
europepmc +1 more source
A two‐terminal IGZO memtransistor enabling low‐voltage operation is demonstrated through coupling enhancement. By utilizing drain‐induced Fowler–Nordheim tunneling, the device achieves programming below 10 V with a large memory window and high on/off ratio.
Junhyeong Park +5 more
wiley +1 more source
Decoupled Spin Hall Voltage Readout for Domain‐Wall Synapses in Embedded Neuromorphic Computing
Magnetic domain‐wall synapses provide nonvolatile multilevel states with linear, symmetric weight updates, making them attractive for neuromorphic hardware. However, DW MTJ readout has magnetoresistance‐limited dynamic range, while anomalous‐Hall readout suffers from weak voltage output and read/write coupling.
Stanislav Sin +2 more
wiley +1 more source
Bio-inspired cognitive robotics vs. embodied AI for socially acceptable, civilized robots. [PDF]
Morasso P.
europepmc +1 more source
Vertical NAND Flash Memory‐Based Bayesian Neural Networks With Intrinsic Gaussian Weight Sampling
Commercial vertical NAND flash memory supports Bayesian neural networks by using paired cells to produce bell‐shaped weight distributions. Read voltage controls their spread without additional memory cells or changes to the array structure. Simulations based on device measurements demonstrate improved uncertainty estimation and accuracy gains of up to ...
Sung‐Ho Park +8 more
wiley +1 more source
Information-Theoretic Framework for Quantum State Purification and Error Correction via Symmetric Subspace Projection. [PDF]
Tang J, Wang MJ.
europepmc +1 more source

