arXiv subjects

Bert Jan Offrein

Publications and source records attributed to Bert Jan Offrein.

13 recordsLinked to original sources

arXiv2026

Update Disturbance-Resilient Analog ReRAM Crossbar Arrays for In-Memory Deep Learning Accelerators

Resistive memory (ReRAM) technologies with crossbar array architectures hold significant potential for analog AI accelerator hardware, enabling both in-memory inference and training. Recent developments have successfully demonstrated inference acceleration by offloading compute-heavy training workloads to off-chip digital processors. However, in-memory acceleration of training algorithms is crucial for more sustainable and power-efficient AI, but still in an early stage of research. This study addresses in-memory training acceleration using analog ReRAM arrays, focusing on a key challenge during fully parallel weight updates: disturbances of the weight values in cross-point devices. A ReRAM device solution is presented on 350 nm silicon technology, utilizing a resistive switching conductive metal oxide (CMO) formed on a nanoscale conductive filament within a HfOx layer. The devices not only exhibit 60 ns fast, non-volatile analog switching, but also demonstrates outstanding resilience to update disturbances, enduring over 100k pulses. The disturbance tolerance of the ReRAM is analyzed using COMSOL Multiphysics simulations, modeling the filament-induced thermoelectric energy concentration that results in a highly nonlinear device responses to input voltage amplitudes. Disturbance-free parallel weight mapping is also demonstrated on the back-end-of-line integrated ReRAM array chip. Finally, comprehensive hardware-aware neural network simulations validate the potential of our ReRAM for in-memory deep learning accelerators capable of fully parallel weight updates.

cs.ET↗

arXiv2026

Study of Resistive Switching Dynamics and Memory States Equilibria in Analog Filamentary Conductive-Metal-Oxide/HfOx ReRAM via Compact Modeling

Resistive Random Access Memory (ReRAM) devices offer a promising solution for next-generation non-volatile memory and neuromorphic computing systems. Yet, existing compact models fail to capture analog resistive switching behavior of ReRAM devices. This work presents an advanced physics-based compact model for analog filamentary Conductive-Metal-Oxide (CMO)/HfOx ReRAM, capable of reproducing switching characteristics over a broad range of operating conditions. Compared to the state-of-the-art, the model extends the dynamic interplay between ion migration and electron hopping, while also accounting for parasitic resistive elements. Simulations of various voltage inputs are tested to reproduce quasi-static I-V curves, SET switching kinetics under single-pulse programming conditions, and analog accumulative conductance modulation upon bipolar identical pulse streams. Additional simulations reveal the physical criterion underlying the stabilization of the CMO/HfOx-based ReRAM memory state around the equilibrium point, namely symmetry point, under pulsing conditions when a fading memory mechanism emerges. Building upon the evidence of such equilibrium stabilization under pulsing and quasi-static conditions, a procedure is established to visualize and map equilibrium memory states across different input domains. The physical model supports design optimization of switching behavior for analog neuromorphic systems and non-volatile memory architectures. It also enables accurate integrated circuit simulations with CMO/HfOx-based ReRAM technology.

cs.ET↗

arXiv2026

Analog Weight Update Rule in Ferroelectric Hafnia, using pico-Joule Programming Pulses

In an effort to compete with the brain's efficiency at processing information, neuromorphic hardware combines artificial synapses and neurons using mixed-signal circuits and emerging memories. In ferroelectric resistive weights, the strength of the synaptic connection between two neurons is stored in the device conductance. During learning, programming pulses are applied to the synaptic weight, which reconfigures the ferroelectric domains and adjusts the conductance. One strategy to lower the energy cost during the training phase is to lower the duration of the programming pulses. However, the latter cannot be shorter than the self-loading time of the resistive weights, limited by intrinsic parasitics in the circuits. In this work, ferroelectric resistive weights are fabricated using a process compatible with CMOS Back-End-Of-Line integration, based on hafnia/zirconia nanolaminates. By laterally scaling the device area under 100 $μ$m$^2$, the self-loading time becomes sufficiently short to enable 20 ns programming, which corresponds to a maximum of 3 picoJoules per pulse. Further, in this work, the weight update rule with 20 ns pulses is experimentally measured not only for different amplitudes but also for different initial conductance states. We find that the final weight is determined by the pulse amplitude, independent of the initial weight value.

cs.ET↗

arXiv2025

Unsupervised local learning based on voltage-dependent synaptic plasticity for resistive and ferroelectric synapses

The deployment of AI on edge computing devices faces significant challenges related to energy consumption and functionality. These devices could greatly benefit from brain-inspired learning mechanisms, allowing for real-time adaptation while using low-power. In-memory computing with nanoscale resistive memories may play a crucial role in enabling the execution of AI workloads on these edge devices. In this study, we introduce voltage-dependent synaptic plasticity (VDSP) as an efficient approach for unsupervised and local learning in memristive synapses based on Hebbian principles. This method enables online learning without requiring complex pulse-shaping circuits typically necessary for spike-timing-dependent plasticity (STDP). We show how VDSP can be advantageously adapted to three types of memristive devices (TiO$_2$, HfO$_2$-based metal-oxide filamentary synapses, and HfZrO$_4$-based ferroelectric tunnel junctions (FTJ)) with disctinctive switching characteristics. System-level simulations of spiking neural networks incorporating these devices were conducted to validate unsupervised learning on MNIST-based pattern recognition tasks, achieving state-of-the-art performance. The results demonstrated over 83% accuracy across all devices using 200 neurons. Additionally, we assessed the impact of device variability, such as switching thresholds and ratios between high and low resistance state levels, and proposed mitigation strategies to enhance robustness.

cs.NE↗

arXiv2025

Analytical Modelling of the Transport in Analog Filamentary Conductive-Metal-Oxide/HfOx ReRAM Devices

The recent co-optimization of memristive technologies and programming algorithms enabled neural networks training with in-memory computing systems. In this context, novel analog filamentary conductive-metal-oxide (CMO)/HfOx redox-based resistive switching memory (ReRAM) represents a key technology. Despite device performance enhancements reported in literature, the underlying mechanism behind resistive switching is not fully understood. This work presents the first physics-based analytical model of the current transport and of the resistive switching in these devices. As a case study, analog TaOx/HfOx ReRAM devices are considered. The current transport is explained by a trap-to-trap tunneling process, and the resistive switching by a modulation of the defect density within the sub-band of the TaOx that behaves as electric field and temperature confinement layer. The local temperature and electric field distributions are derived from the solution of the electric and heat transport equations in a 3D finite element ReRAM model. The intermediate resistive states are described as a gradual modulation of the TaOx defect density, which results in a variation of its electrical conductivity. The drift-dynamics of ions during the resistive switching is analytically described, allowing the estimation of defect migration energies in the TaOx layer. Moreover, the role of the electro-thermal properties of the CMO layer is unveiled. The proposed analytical model accurately describes the experimental switching characteristic of analog TaOx/HfOx ReRAM devices, increasing the physical understanding and providing the equations necessary for circuit simulations incorporating this technology.

cs.ET↗

arXiv2025

Energy-convergence trade off for the training of neural networks on bio-inspired hardware

The increasing deployment of wearable sensors and implantable devices is shifting AI processing demands to the extreme edge, necessitating ultra-low power for continuous operation. Inspired by the brain, emerging memristive devices promise to accelerate neural network training by eliminating costly data transfers between compute and memory. Though, balancing performance and energy efficiency remains a challenge. We investigate ferroelectric synaptic devices based on HfO2/ZrO2 superlattices and feed their experimentally measured weight updates into hardware-aware neural network simulations. Across pulse widths from 20 ns to 0.2 ms, shorter pulses lower per-update energy but require more training epochs while still reducing total energy without sacrificing accuracy. Classification accuracy using plain stochastic gradient descent (SGD) is diminished compared to mixed-precision SGD. We analyze the causes and propose a ``symmetry point shifting'' technique, addressing asymmetric updates and restoring accuracy. These results highlight a trade-off among accuracy, convergence speed, and energy use, showing that short-pulse programming with tailored training significantly enhances on-chip learning efficiency.

cs.ET↗

arXiv2025

All-in-One Analog AI Hardware: On-Chip Training and Inference with Conductive-Metal-Oxide/HfOx ReRAM Devices

Analog in-memory computing is an emerging paradigm designed to efficiently accelerate deep neural network workloads. Recent advancements have focused on either inference or training acceleration. However, a unified analog in-memory technology platform-capable of on-chip training, weight retention, and long-term inference acceleration-has yet to be reported. This work presents an all-in-one analog AI accelerator, combining these capabilities to enable energy-efficient, continuously adaptable AI systems. The platform leverages an array of analog filamentary conductive-metal-oxide (CMO)/HfOx resistive switching memory cells (ReRAM) integrated into the back-end-of-line (BEOL). The array demonstrates reliable resistive switching with voltage amplitudes below 1.5V, compatible with advanced technology nodes. The array multi-bit capability (over 32 stable states) and low programming noise (down to 10nS) enable a nearly ideal weight transfer process, more than an order of magnitude better than other memristive technologies. Inference performance is validated through matrix-vector multiplication simulations on a 64x64 array, achieving a root-mean-square error improvement by a factor of 20 at 1 second and 3 at 10 years after programming, compared to state-of-the-art. Training accuracy closely matching the software equivalent is achieved across different datasets. The CMO/HfOx ReRAM technology lays the foundation for efficient analog systems accelerating both inference and training in deep neural networks.

cs.ET↗

arXiv2025

Hardware Implementation of Ring Oscillator Networks Coupled by BEOL Integrated ReRAM for Associative Memory Tasks

We demonstrate the first hardware implementation of an oscillatory neural network (ONN) utilizing resistive memory (ReRAM) for coupling elements. A ReRAM crossbar array chip, integrated into the Back End of Line (BEOL) of CMOS technology, is leveraged to establish dense coupling elements between oscillator neurons, allowing phase-encoded analog information to be processed in-memory. We also realize an ONN architecture design with the coupling ReRAM array. To validate the architecture experimentally, we present a conductive metal oxide (CMO)/HfOx ReRAM array chip integrated with a 2-by-2 ring oscillator-based network. The system successfully retrieves patterns through correct binary phase locking. This proof of concept underscores the potential of ReRAM technology for large-scale, integrated ONNs.

cs.ET↗

arXiv2025

Edge Training and Inference with Analog ReRAM Technology for Hand Gesture Recognition

Tactile hand gesture recognition is a crucial task for user control in the automotive sector, where Human-Machine Interactions (HMI) demand low latency and high energy efficiency. This study addresses the challenges of power-constrained edge training and inference by utilizing analog Resistive Random Access Memory (ReRAM) technology in conjunction with a real tactile hand gesture dataset. By optimizing the input space through a feature engineering strategy, we avoid relying on large-scale crossbar arrays, making the system more suitable for edge deployment. Through realistic hardware-aware simulations that account for device non-idealities derived from experimental data, we demonstrate the functionalities of our analog ReRAM-based analog in-memory computing for on-chip training, utilizing the state-of-the-art Tiki-Taka algorithm. Furthermore, we validate the classification accuracy of approximately 91.4% for post-deployment inference of hand gestures. The results highlight the potential of analog ReRAM technology and crossbar architecture with fully parallelized matrix computations for real-time HMI systems at the Edge.

eess.SP↗

arXiv2023

A BEOL Compatible, 2-Terminals, Ferroelectric Analog Non-Volatile Memory

A Ferroelectric Analog Non-Volatile Memory based on a WOx electrode and ferroelectric HfZrO$_4$ layer is fabricated at a low thermal budget (~375$^\circ$C), enabling BEOL processes and CMOS integration. The devices show suitable properties for integration in crossbar arrays and neural network inference: analog potentiation/depression with constant field or constant pulse width schemes, cycle to cycle and device to device variation <10%, ON/OFF ratio up to 10 and good linearity. The physical mechanisms behind the resistive switching and conduction mechanisms are discussed.

cs.AR↗

arXiv2023

High-Conductance, Ohmic-like HfZrO$_4$ Ferroelectric Memristor

The persistent and switchable polarization of ferroelectric materials based on HfO$_2$-based ferroelectric compounds, compatible with large-scale integration, are attractive synaptic elements for neuromorphic computing. To achieve a record current density of 0.01 A/cm$^2$ (at a read voltage of 80 mV) as well as ideal memristive behavior (linear current-voltage relation and analog resistive switching), devices based on an ultra-thin (2.7 nm thick), polycrystalline HfZrO$_4$ ferroelectric layer are fabricated by Atomic Layer Deposition. The use of a semiconducting oxide interlayer (WO$_{x<3}$) at one of the interfaces, induces an asymmetric energy profile upon ferroelectric polarization reversal and thus the long-term potentiation / depression (conductance increase / decrease) of interest. Moreover, it favors the stable retention of both the low and the high resistive states. Thanks to the low operating voltage (<3.5 V), programming requires less than 10${^-12}$ J for 20 ns long pulses. Remarkably, the memristors show no wake-up or fatigue effect.

cs.AR↗

arXiv2023

A Back-End-Of-Line Compatible, Ferroelectric Analog Non-Volatile Memory

A Ferroelectric Analog Non-Volatile Memory based on a WOx electrode and ferroelectric HfZrO4 layer is fabricated at a low thermal budget (~375C), enabling BEOL processes and CMOS integration. The devices show suitable properties for integration in crossbar arrays and neural network inference: analog potentiation/depression with constant field or constant pulse width schemes, cycle to cycle and device to device variation <10%, ON/OFF ratio up to 10 and good linearity. The physical mechanisms behind the resistive switching and conduction mechanisms are discussed.

cs.AR↗

arXiv2020

A back-end, CMOS compatible ferroelectric Field Effect Transistor for synaptic weights

Neuromorphic computing architectures enable the dense co-location of memory and processing elements within a single circuit. This co-location removes the communication bottleneck of transferring data between separate memory and computing units as in standard von Neuman architectures for data-critical applications including machine learning. The essential building blocks of neuromorphic systems are non-volatile synaptic elements such as memristors. Key memristor properties include a suitable non-volatile resistance range, continuous linear resistance modulation and symmetric switching. In this work, we demonstrate voltage-controlled, symmetric and analog potentiation and depression of a ferroelectric Hf$_{57}$Zr$_{43}$O$_{2}$ (HZO) field effect transistor (FeFET) with good linearity. Our FeFET operates with a low writing energy (fJ) and fast programming time (40 ns). Retention measurements have been done over 4-bits depth with low noise (1%) in the tungsten oxide (WO$_{x}$) read out channel. By adjusting the channel thickness from 15nm to 8nm, the on/off ratio of the FeFET can be engineered from 1% to 200% with an on-resistance ideally >100 kOhm, depending on the channel geometry. The device concept is using earth-abundant materials, and is compatible with a back end of line (BEOL) integration into complementary metal-oxidesemiconductor (CMOS) processes. It has therefore a great potential for the fabrication of high density, large-scale integrated arrays of artificial analog synapses.

cs.ET↗