Search

Abstract Resistive Random Access Memories (ReRAM) arrays provides a promising basement to deploy neural network accelerators based on near or in memory computing. However most popular accelerators rely on Ohm’s and Kirchhoff’s laws to achieve multiply and accumulate, and thus are prone to ReRAM variability and voltage drop in the memory array, and thus need sophisticated readout circuits. Here we propose a robust binary neural network, based on fully differential capacitive neurons and ReRAM synapses, used in a resistive bridge fashion. We fabricated a network layer with up to 23 inputs that we extrapolated to large numbers of inputs through simulation. Defining proper programming and reading conditions, we demonstrate the high resilience of this solution with a minimal accuracy drop, compared to a software baseline, on image classification tasks. Moreover, our solution can achieve a peak energy efficiency, comparable with the state of the art, when projected to a 22 nanometer technology.

Abstract Safety-critical sensory applications, like medical diagnosis, demand accurate decisions from limited, noisy data. Bayesian neural networks excel at such tasks, offering predictive uncertainty assessment. However, because of their probabilistic nature, they are computationally intensive. An innovative solution utilizes memristors’ inherent probabilistic nature to implement Bayesian neural networks. However, when using memristors, statistical effects follow the laws of device physics, whereas in Bayesian neural networks, those effects can take arbitrary shapes. This work overcome this difficulty by adopting a variational inference training augmented by a “technological loss”, incorporating memristor physics. This technique enabled programming a Bayesian neural network on 75 crossbar arrays of 1,024 memristors, incorporating CMOS periphery for in-memory computing. The experimental neural network classified heartbeats with high accuracy, and estimated the certainty of its predictions. The results reveal orders-of-magnitude improvement in inference energy efficiency compared to a microcontroller or an embedded graphics processing unit performing the same task.

Crossbars of resistive memories, or memristors, provide a road to reduce the energy consumption of artificial neural networks, by naturally implementing multiply accumulate operations, their most basic calculations. However, a major challenge of implementing robust hardware neural networks is the conductance instability over time of resistive memories, due to the local recombination of oxygen vacancies. This effect causes resistive memory‐based neural networks to rapidly lose accuracy, an issue that is sometimes overlooked. Herein, this conductance instability issue is shown, which can be avoided without changing the material stack of the resistive memory by exploiting an original programming strategy. This technique relies on program‐and‐verify loops with appropriately chosen wait times and ensures that the resistive memories are programmed into states with stable filaments. To test the strategy, a 32 × 32 in‐memory computing system, fabricated in a hybrid complementary metal‐oxide‐semiconductor (CMOS)/hafnium oxide technology, is programmed to classify heart arrhythmia from electrocardiogram. When the resistive memories are programmed conventionally, the system loses accuracy within hours. In contrast, when using this technique, the system maintains an accuracy of 95% over more than 2 months. These results highlight the potential of resistive memory for the implementation of low‐power neural networks with long‐term stability.

Abstract Neuroscience findings emphasize the role of dendritic branching in neocortical pyramidal neurons for non-linear computations and signal processing. Dendritic branches facilitate temporal feature detection via synaptic delays that enable coincidence detection (CD) mechanisms. Spiking neural networks highlight the significance of delays for spatio-temporal pattern recognition in feed-forward networks, eliminating the need for recurrent structures. Here, we introduce DenRAM, a novel analog electronic feed-forward spiking neural network with dendritic compartments. Utilizing 130 nm technology integrated with resistive RAM (RRAM), DenRAM incorporates both delays and synaptic weights. By configuring RRAMs to emulate bio-realistic delays and exploiting their heterogeneity, DenRAM mimics synaptic delays and efficiently performs CD for pattern recognition. Hardware-aware simulations on temporal benchmarks show DenRAM’s robustness against hardware noise, and its higher accuracy over recurrent networks. DenRAM advances temporal processing in neuromorphic computing, optimizes memory usage, and marks progress in low-power, real-time signal processing

Continuous renal replacement therapies (CRRTs) and sequential extracorporeal blood purification (EBP) therapies can be used in patients with severe COVID-19 disease to support kidney failure and restore immune homeostasis. EBP prescription should be based on the patient’s clinical needs and frequently re-evaluated during the intensive care unit (ICU) stay. Personalization of treatment at the bedside plays a fundamental role for patient recovery. This aim can be simplified by using both clinical and molecular data collected from a patient-individualized web registry. In this case report, we describe how we apply a sequential approach to EBP therapies following the rapid evolution of a critically ill COVID-19 patient with acute kidney injury. We show patient strategies and outcomes using bedside data from a registry-based method for the routine use of EBP. We explain the choice of specific hemofilter prescription, also focusing on dose and anticoagulation strategies. We describe the difficulties, uncertainties, and mistakes made during EBP prescription. Furthermore, we discuss the causes and workable solutions that can be adopted by the ICU physician for a better EBP prescription, considering the current lack of well-recognized indications.

The real-world object localization application needs a low-latency and power efficient computing system. Here, Moro et al. demonstrate a neuromorphic in-memory event driven system, inspired by the barn owl’s neuroanatomy, which is orders of magnitude more energy efficient than microcontrollers.