npj Unconventional Computing

Emerging processing-in-memory (PIM) architectures using memristors and analog computing face reliability issues from device non-idealities and noise. While error-correcting codes (ECCs) are vital, existing methods suffer from discontinuity and inefficiency. We propose a non-binary low-density parity-check (NB-LDPC) code over Galois field (GF) to address this. The design employs a 1024-symbol info…

Real-time robotic systems require advanced perception and action capability. However, the main bottleneck in current autonomous systems is the trade-off between computational capability, energy efficiency, and model determinism. World modeling, a key objective of many robotic systems, commonly uses occupancy grid mapping (OGM) as the first step towards building an end-to-end robotic system. OGM d…

Abstract We present a low-power analog CMOS reservoir computing chip that employs a simple ring topology for unidirectional nonlinear analog processing. Each node integrates a sample-and-hold circuit operating at 1 kHz, enabling compact and energy-efficient implementation in standard CMOS. The fabricated chip was evaluated on standard benchmarks, achieving a linear memory capacity of 13.4 and inf…

Harnessing the analog capacity of quantum processors at the algorithmic level is key to solving computationally hard problems. Neutral atoms offer analog capabilities supporting hundreds of qubits, but state-of-the-art adiabatic protocol struggles with nonadiabatic errors, restricting scalability due to finite coherence times. To address this, we propose and experimentally demonstrate a tailored …

Abstract Combinatorial optimization (CO) underpins critical applications in science and engineering, ranging from logistics to electronic design automation. A classic example of CO is the NP-complete Traveling Salesman Problem (TSP). Finding exact solutions for large-scale TSP instances remains computationally intractable; on von Neumann architectures, such solvers are constrained by the memory w…

The pursuit of high-performance and energy-efficient computing for data-intensive algorithms such as deep neural networks (DNN) opens up exciting opportunities for emerging non-volatile memories (NVM). Particularly, implementing such non-volatile memory units in crossbar arrays as weight matrix storage can provide highly parallel and efficient means of processing matrix-vector multiplications, pr…

Classical mean-variance optimization is powerful in theory but fragile in practice, often producing highly concentrated, high-turnover portfolios. Naive equal-weight (1/N) portfolios are more robust but largely ignore cross-sectional information. We propose a quantum stochastic walk (QSW) framework that embeds assets in a weighted graph and derives portfolio weights from the stationary distributi…

Physical reservoir computing leverages the intrinsic history-dependence and nonlinearity of hardware to encode spatiotemporal signals directly at the sensor level, enabling low-latency processing of dynamic inputs. Encoding fidelity depends on the separability of multi-state outputs, yet in practice it is often hampered by empirically chosen, suboptimal operating conditions. Here, we apply Bayesi…

Second-order training methods have better convergence properties than gradient descent but are rarely used in practice for large-scale training due to their computational overhead. This can be viewed as a hardware limitation (imposed by digital computers). Here, we show that natural gradient descent (NGD), a second-order method, can have a similar computational complexity per iteration to a first…

This work presents a hardware-algorithm co-designed framework for neuromorphic computing, enabling efficient supervised learning in spike-based neural architectures. First, synaptic updates are reformulated as low-rank outer products of forward spike vectors and backward error gradients via singular value decomposition (SVD), enabling direct parallelization on 1T1R arrays. Second, a stochastic co…

Abstract In the present study, a modified Leaky-Integrate and Fire (LIF) neuron model termed a Hybrid Spiking Neuron (HSN) is proposed and introduced as a physics-based meta-learning solver for applications in engineering mechanics. Unlike LIF neurons, HSNs produce a real-valued spiking signal. In each time step, the activation function determines whether the neuron is active and outputs its real…

Abstract Artificial intelligence (AI) is accelerating the evolution of robotics from task-specific automation to general-purpose autonomy, enabling robots to perform high-level tasks in unstructured and dynamic environments. One of the key enablers in this evolution is the integration of AI with robotic vision systems, which provide accurate perception and contextual interpretation of complex sur…

This work presents an electrically tunable physical reservoir computing (RC) system using two-terminal Pt/HfO2/TiN memristors. The device’s intrinsic relaxation time is effectively increased by over two orders of magnitude by modulating the base voltage between data pulses, enabling robust temporal data encoding for pulse intervals ranging from 0.14 to 100 ms. Enhanced separability is experimenta…

Neuromorphic systems constructed from biomolecular materials offer rich dynamics and memory properties, enabling low-power, brain-like signal processing and computing. Unlike many neuromorphic devices, biomolecular materials and systems emulate the nanoscale architectures and physical mechanisms of biological synapses and neurons, which are responsible for the brain’s complex computing functions …

Artificial intelligence technology transforms traditional sensors from passive data collectors into active computing nodes, performing data processing at the edge. This paradigm shift toward in- and near-sensor computing mitigates inherent inefficiencies associated with data traversal between sensing, memory, and processing units. We introduce emerging device technologies, circuit architectures, …

This paper investigates the combined potential of neuromorphic and edge computing to develop a flexible machine learning (ML) system designed for processing data from dynamic vision sensors. We build and train hybrid models that integrate spiking neural networks (SNNs) and artificial neural networks (ANNs) using the PyTorch and Lava frameworks. We explore the effects of quantization on ANN models…