arXiv subjects

Abhronil Sengupta

Publications and source records attributed to Abhronil Sengupta.

At least 19 recordsLinked to original sources

arXiv2026

STAR: Astrocyte-Inspired State-Augmented Repair for Supervised Memristive AI Hardware Systems

Memristive crossbar arrays have emerged as a promising platform for efficient on-chip learning, enabling local learning rules such as Equilibrium Propagation (EP) to be realized without the memory overhead of conventional backpropagation. However, as these devices age, permanent stuck-at (SA) faults accumulate at individual synaptic elements, irreversibly corrupting the stored weights and degrading model performance in ways that in-situ retraining alone cannot address. Existing mitigation strategies either rely on hardware redundancy at significant area and power cost, or require explicit fault localization that is impractical to perform continuously on-chip. This paper adopts a brain-inspired fault recovery route motivated by self-repair functionalities enabled by astrocytes -- a type of glial cell. We couple a computational neuroscience study of astrocytic neuromodulation under permanent synaptic faults with an algorithmic recovery mechanism for faulted crossbar-mapped networks. Our computational neuroscience study characterizes how astrocytes modulate surviving synapses in a bidirectional EP-specific network setting under both SA-0 and high-conductance stuck-at fault conditions, revealing that recovery operates at the neural population level. Motivated by this observation, we propose a retraining-based repair mechanism that augments EP with an additional repair nudge anchored to pre-fault activation targets of the healthy network, encouraging surviving weights to collectively reconstruct pre-fault internal representations without any information of which weights are faulty. To our knowledge, STAR is the first astrocyte-inspired repair mechanism to operate under a supervised local learning rule and to scale to convolutional networks.

cs.ET↗

arXiv2026

Neuromorphic Reinforcement Learning for Quadruped Locomotion Control on Uneven Terrain

Reinforcement learning (RL) has enabled robust quadruped locomotion over complex terrain, but most learned controllers are trained offline with backpropagation in massively parallel simulation and deployed as fixed policies, limiting adaptation to terrain variation, payload changes, actuator wear, and other real-world conditions under onboard power constraints. Local learning provides a potential path toward energy-aware on-robot adaptation by replacing global backpropagation graphs with updates driven by local neural states, making the learning rule more compatible with neuromorphic and in-memory computing substrates. This work proposes an equilibrium-propagation (EP)-based proximal policy optimization (PPO) framework for uneven-terrain quadruped locomotion. The controller combines a bio-inspired central pattern generator (CPG) policy with a residual postural adjustment policy, while replacing conventional backpropagation-trained policy and value networks with EP-enabled local learning. To train stochastic continuous-control policies with EP, we derive an EP-compatible PPO output-nudging signal and introduce a two-sided ratio clipping mechanism that stabilizes policy updates during relaxation. Experiments on a 12-DoF A1 quadruped show that the proposed controller achieves stable policy convergence in a two-stage uneven terrain locomotion task. Its locomotion performance is comparable to a backpropagation-trained PPO baseline in success rate, velocity tracking, actuator power, and body stability, while improving GPU memory efficiency by 4.3$\times$ compared with backpropagation through time (BPTT). These results suggest that local equilibrium-based learning can support high-dimensional embodied locomotion and provide an algorithmic foundation for low-power on-robot adaptation and fine-tuning.

cs.NE↗

arXiv2026

Bayesian Optimization of Crossbar-Based Compute-In-Memory System Design for Efficient DNN Inference

Leveraging the high density and energy efficiency of Compute-In-Memory (CIM) crossbar-based Deep Neural Network (DNN) accelerators requires optimal Design Space Exploration (DSE), which becomes increasingly challenging as complex models for advanced AI workloads expand the highly non-convex design space. Among existing DSE approaches, multi-objective Bayesian Optimization (BO) is promising, as it explores high-quality design solutions while querying costly CIM simulators selectively. In this work, we propose a multi-objective BO framework that holistically co-optimizes hardware and algorithm parameters of a CIM crossbar-based hardware accelerator for various DNN inference tasks. Depending on NN model depth, our framework handles high-dimensional design spaces (with $26$ and $50$ dimensions) and extremely large search complexities on the order of $O(10^{12})$ and $O(10^{27})$ for VGG8/CIFAR-10 and VGG16/Tiny-ImageNet-200. Our method attains $91.72 \%$ and $57.2 \%$ accuracy, respectively, comparable to baseline designs, while improving chip area ($65.52 \%$ and $50.7 \%$), read latency ($9.52 \%$ and $13.27 \%$), read dynamic energy ($31.23 \%$ and $52.07 \%$) and increasing memory utilization ($13.41 \%$ and $2.67 \%$).

cs.ET↗

arXiv2026

NeuroAI and Beyond: Bridging Between Advances in Neuroscience and ArtificialIntelligence

Neuroscience and Artificial Intelligence (AI) have made impressive progress in recent years but remain only loosely interconnected. Based on a workshop convened by the National Science Foundation in August 2025, we identify three fundamental capability gaps in current AI: the inability to interact with the physical world, inadequate learning that produces brittle systems, and unsustainable energy and data inefficiency. We describe the neuroscience principles that address each: co-design of body and controller, prediction through interaction, multi-scale learning with neuromodulatory control, hierarchical distributed architectures, and sparse event-driven computation. We present a research roadmap organized around these principles at near, mid, and long-term horizons. We argue that realizing this program requires a new generation of researchers trained across the boundary between neuroscience and engineering, and describe the institutional conditions: interdisciplinary training, hardware access, community standards, and ethics, needed to support them. We conclude that NeuroAI, neuroscience-informed artificial intelligence, has the potential to overcome limitations of current AI while deepening our understanding of biological neural computation.

q-bio.NC↗

arXiv2026

Trilinear Compute-in-Memory Architecture for Energy-Efficient Transformer Acceleration

Self-attention in Transformers generates dynamic operands that force conventional Compute-in-Memory (CIM) accelerators into costly non-volatile memory (NVM) reprogramming cycles, degrading throughput and stressing device endurance. Existing solutions either reduce but retain NVM writes through matrix decomposition or sparsity, or move attention computation to digital CMOS at the expense of NVM density. We present TrilinearCIM, a Double-Gate FeFET (DG-FeFET)-based architecture that uses back-gate modulation to realize a three-operand multiply-accumulate primitive for in-memory attention computation without dynamic ferroelectric reprogramming. Evaluated on BERT-base (GLUE) and ViT-base (ImageNet and CIFAR), TrilinearCIM outperforms conventional CIM on seven of nine GLUE tasks while achieving up to 46.6\% energy reduction and 20.4\% latency improvement over conventional FeFET CIM at 37.3\% area overhead. To our knowledge, this is the first architecture to perform complete Transformer attention computation exclusively in NVM cores without runtime reprogramming.

cs.AR↗

arXiv2026

RMAAT: Astrocyte-Inspired Memory Compression and Replay for Efficient Long-Context Transformers

The quadratic complexity of self-attention mechanism presents a significant impediment to applying Transformer models to long sequences. This work explores computational principles derived from astrocytes-glial cells critical for biological memory and synaptic modulation-as a complementary approach to conventional architectural modifications for efficient self-attention. We introduce the Recurrent Memory Augmented Astromorphic Transformer (RMAAT), an architecture integrating abstracted astrocyte functionalities. RMAAT employs a recurrent, segment-based processing strategy where persistent memory tokens propagate contextual information. An adaptive compression mechanism, governed by a novel retention factor derived from simulated astrocyte long-term plasticity (LTP), modulates these tokens. Attention within segments utilizes an efficient, linear-complexity mechanism inspired by astrocyte short-term plasticity (STP). Training is performed using Astrocytic Memory Replay Backpropagation (AMRB), a novel algorithm designed for memory efficiency in recurrent networks. Evaluations on the Long Range Arena (LRA) benchmark demonstrate RMAAT's competitive accuracy and substantial improvements in computational and memory efficiency, indicating the potential of incorporating astrocyte-inspired dynamics into scalable sequence models.

cs.NE↗

arXiv2025

GRASP: GRouped Activation Shared Parameterization for Parameter-Efficient Fine-Tuning and Robust Inference of Transformers

Parameter-efficient fine-tuning (PEFT) provides a scalable alternative to full-model adaptation by updating only a small subset of parameters in large pre-trained models. We introduce GRASP - GRouped Activation Shared Parameterization - a lightweight PEFT framework that partitions the D-dimensional token representations of selected layers into K << D groups and learns a shared scaling and shifting vector for each group. This grouped modulation reduces the number of trainable parameters significantly while preserving the ability of the model to learn task-specific features. Building on this formulation, we further propose StochGRASP, which learns Gaussian distributions as perturbations to the pre-trained weights rather than deterministic values. This probabilistic parameterization along with a noise-aware loss function formulation enables modelling hardware-level variability in programmed weights and significantly improves robustness under non-ideal inference conditions-an important requirement for deployment on edge-based emerging AI hardware. Across GLUE (RoBERTa-base & RoBERTa-large) and E2E NLG (GPT-2 Medium), GRASP matches or exceeds the performance of established PEFT methods while achieving an order of magnitude reduction in trainable parameters compared to LoRA and BitFit. Under varying levels of noise, StochGRASP consistently outperforms deterministic variants, demonstrating its suitability for energy-efficient and noise-prone hardware platforms.

cs.LG↗

arXiv2025

StochEP: Stochastic Equilibrium Propagation for Spiking Convergent Recurrent Neural Networks

Spiking Neural Networks (SNNs) promise energy-efficient, sparse, biologically inspired computation. Training them with Backpropagation Through Time (BPTT) and surrogate gradients achieves strong performance but remains biologically implausible. Equilibrium Propagation (EP) provides a more local and biologically grounded alternative. However, existing EP frameworks, primarily based on deterministic neurons, either require complex mechanisms to handle discontinuities in spiking dynamics or fail to scale beyond simple visual tasks. Inspired by the stochastic nature of biological spiking mechanism and recent hardware trends, we propose a stochastic EP framework that integrates probabilistic spiking neurons into the EP paradigm. This formulation smoothens the optimization landscape, stabilizes training, and enables scalable learning in deep convolutional spiking convergent recurrent neural networks (CRNNs). We provide theoretical guarantees showing that the proposed stochastic EP dynamics approximate deterministic EP under mean-field theory, thereby inheriting its underlying theoretical guarantees. The proposed framework narrows the gap to both BPTT-trained SNNs and EP-trained non-spiking CRNNs in vision benchmarks while preserving locality, highlighting stochastic EP as a promising direction for neuromorphic and on-chip learning.

cs.ET↗

arXiv2025

Dendritic Computing with Multi-Gate Ferroelectric Field-Effect Transistors

Although inspired by neuronal systems in the brain, artificial neural networks generally employ point-neurons, which offer far less computational complexity than their biological counterparts. Neurons have dendritic arbors that connect to different sets of synapses and offer local non-linear accumulation - playing a pivotal role in processing and learning. Inspired by this, we propose a novel neuron design based on a multi-gate ferroelectric field-effect transistor that mimics dendrites. It leverages ferroelectric nonlinearity for local computations within dendritic branches, while utilizing the transistor action to generate the final neuronal output. The branched architecture paves the way for utilizing smaller crossbar arrays in hardware integration, leading to greater efficiency. Using an experimentally calibrated device-circuit-algorithm co-simulation framework, we demonstrate that networks incorporating our dendritic neurons achieve superior performance in comparison to much larger networks without dendrites ($\sim$17$\times$ fewer trainable weight parameters). These findings suggest that dendritic hardware can significantly improve computational efficiency, and learning capacity of neuromorphic systems optimized for edge applications.

cs.ET↗

arXiv2025

Astrocyte Regulated Neuromorphic Central Pattern Generator Control of Legged Robotic Locomotion

Neuromorphic computing systems, where information is transmitted through action potentials in a bio-plausible fashion, is gaining increasing interest due to its promise of low-power event-driven computing. Application of neuromorphic computing in robotic locomotion research have largely focused on Central Pattern Generators (CPGs) for bionics robotic control algorithms - inspired from neural circuits governing the collaboration of the limb muscles in animal movement. Implementation of artificial CPGs on neuromorphic hardware platforms can potentially enable adaptive and energy-efficient edge robotics applications in resource constrained environments. However, underlying rewiring mechanisms in CPG for gait emergence process is not well understood. This work addresses the missing gap in literature pertaining to CPG plasticity and underscores the critical homeostatic functionality of astrocytes - a cellular component in the brain that is believed to play a major role in multiple brain functions. This paper introduces an astrocyte regulated Spiking Neural Network (SNN)-based CPG for learning locomotion gait through Reward-Modulated STDP for quadruped robots, where the astrocytes help build inhibitory connections among the artificial motor neurons in different limbs. The SNN-based CPG is simulated on a multi-object physics simulation platform resulting in the emergence of a trotting gait while running the robot on flat ground. $23.3\times$ computational power savings is observed in comparison to a state-of-the-art reinforcement learning based robot control algorithm. Such a neuroscience-algorithm co-design approach can potentially enable a quantum leap in the functionality of neuromorphic systems incorporating glial cell functionality.

cs.NE↗

arXiv2025

On the Adversarial Robustness of Spiking Neural Networks Trained by Local Learning

Recent research has shown the vulnerability of Spiking Neural Networks (SNNs) under adversarial examples that are nearly indistinguishable from clean data in the context of frame-based and event-based information. The majority of these studies are constrained in generating adversarial examples using Backpropagation Through Time (BPTT), a gradient-based method which lacks biological plausibility. In contrast, local learning methods, which relax many of BPTT's constraints, remain under-explored in the context of adversarial attacks. To address this problem, we examine adversarial robustness in SNNs through the framework of four types of training algorithms. We provide an in-depth analysis of the ineffectiveness of gradient-based adversarial attacks to generate adversarial instances in this scenario. To overcome these limitations, we introduce a hybrid adversarial attack paradigm that leverages the transferability of adversarial instances. The proposed hybrid approach demonstrates superior performance, outperforming existing adversarial attack methods. Furthermore, the generalizability of the method is assessed under multi-step adversarial attacks, adversarial attacks in black-box FGSM scenarios, and within the non-spiking domain.

cs.LG↗

arXiv2025

Spatio-Temporal Pruning for Compressed Spiking Large Language Models

Large Language Models (LLMs) present significant challenges for deployment in energy-constrained environments due to their large model sizes and high inference latency. Spiking Neural Networks (SNNs), inspired by the sparse event-driven neural processing and energy-efficient information transmission in the brain, offer a promising alternative for achieving low-power computing. Integrating the event-driven efficiency of spiking neurons with the advanced capabilities of LLMs represents a promising direction for power-efficient LLMs. This work specifically delves into the design of compressed spiking LLMs. Here, we revisit spatial and temporal pruning from the perspective of SNNs and propose a novel spatio-temporal pruning framework for Spiking LLMs to optimize computational efficiency while preserving high performance. Our spatial pruning technique reduces the number of active neurons and attention heads, effectively lowering the computational complexity of the model. Meanwhile, temporal pruning minimizes inference latency by dynamically adjusting the number of timesteps required for different layers. By combining these approaches with other compression techniques, we present the first work in the domain of Spiking LLMs to jointly explore spatial pruning, temporal pruning, extreme quantization and knowledge distillation strategies. Extensive experimental evaluation of our proposed framework for SpikingBERT on the large-scale GLUE benchmark demonstrates the efficacy of our approach in terms of computational operations and inference latency. Our approach offers a compelling solution for real-time, low-power natural language processing applications, making Spiking LLMs more practical for deployment on edge devices and in power-constrained settings.

cs.NE↗

arXiv2025

Scalable Equilibrium Propagation via Intermediate Error Signals for Deep Convolutional CRNNs

Equilibrium Propagation (EP) is a biologically inspired local learning rule first proposed for convergent recurrent neural networks (CRNNs), in which synaptic updates depend only on neuron states from two distinct phases. EP estimates gradients that closely align with those computed by Backpropagation Through Time (BPTT) while significantly reducing computational demands, positioning it as a potential candidate for on-chip training in neuromorphic architectures. However, prior studies on EP have been constrained to shallow architectures, as deeper networks suffer from the vanishing gradient problem, leading to convergence difficulties in both energy minimization and gradient computation. To alleviate the vanishing gradient problem in deep EP networks, we propose a novel EP framework that incorporates layer-wise learning signals to provide auxiliary supervision, which enhances the convergence of neuron dynamics. This is the first work to integrate knowledge distillation and local error signals into EP, enabling the training of significantly deeper architectures. Our proposed approach achieves state-of-the-art performance on the CIFAR-10 and CIFAR-100 datasets, showcasing its scalability on deep VGG architectures. These results represent a significant advancement in the scalability of EP, suggesting that intermediate learning signals can extend the practical applicability of EP to deeper architectures.

cs.LG↗

arXiv2025

Neuromorphic Cybersecurity with Semi-supervised Lifelong Learning

Inspired by the brain's hierarchical processing and energy efficiency, this paper presents a Spiking Neural Network (SNN) architecture for lifelong Network Intrusion Detection System (NIDS). The proposed system first employs an efficient static SNN to identify potential intrusions, which then activates an adaptive dynamic SNN responsible for classifying the specific attack type. Mimicking biological adaptation, the dynamic classifier utilizes Grow When Required (GWR)-inspired structural plasticity and a novel Adaptive Spike-Timing-Dependent Plasticity (Ad-STDP) learning rule. These bio-plausible mechanisms enable the network to learn new threats incrementally while preserving existing knowledge. Tested on the UNSW-NB15 benchmark in a continual learning setting, the architecture demonstrates robust adaptation, reduced catastrophic forgetting, and achieves $85.3$\% overall accuracy. Furthermore, simulations using the Intel Lava framework confirm high operational sparsity, highlighting the potential for low-power deployment on neuromorphic hardware.

cs.LG↗

arXiv2025

Benchmarking Spiking Neural Network Learning Methods with Varying Locality

Spiking Neural Networks (SNNs), providing more realistic neuronal dynamics, have been shown to achieve performance comparable to Artificial Neural Networks (ANNs) in several machine learning tasks. Information is processed as spikes within SNNs in an event-based mechanism that significantly reduces energy consumption. However, training SNNs is challenging due to the non-differentiable nature of the spiking mechanism. Traditional approaches, such as Backpropagation Through Time (BPTT), have shown effectiveness but come with additional computational and memory costs and are biologically implausible. In contrast, recent works propose alternative learning methods with varying degrees of locality, demonstrating success in classification tasks. In this work, we show that these methods share similarities during the training process, while they present a trade-off between biological plausibility and performance. Further, given the implicitly recurrent nature of SNNs, this research investigates the influence of the addition of explicit recurrence to SNNs. We experimentally prove that the addition of explicit recurrent weights enhances the robustness of SNNs. We also investigate the performance of local learning methods under gradient and non-gradient-based adversarial attacks.

cs.NE↗

arXiv2025

Delving Deeper Into Astromorphic Transformers

Preliminary attempts at incorporating the critical role of astrocytes - cells that constitute more than 50\% of human brain cells - in brain-inspired neuromorphic computing remain in infancy. This paper seeks to delve deeper into various key aspects of neuron-synapse-astrocyte interactions to mimic self-attention mechanisms in Transformers. The cross-layer perspective explored in this work involves bioplausible modeling of Hebbian and presynaptic plasticities in neuron-astrocyte networks, incorporating effects of non-linearities and feedback along with algorithmic formulations to map the neuron-astrocyte computations to self-attention mechanism and evaluating the impact of incorporating bio-realistic effects from the machine learning application side. Our analysis on sentiment and image classification tasks (IMDB and CIFAR10 datasets) highlights the advantages of Astromorphic Transformers, offering improved accuracy and learning speed. Furthermore, the model demonstrates strong natural language generation capabilities on the WikiText-2 dataset, achieving better perplexity compared to conventional models, thus showcasing enhanced generalization and stability across diverse machine learning tasks.

cs.NE↗

arXiv2025

A Bio-inspired Asymmetric Double-Gate Ferroelectric FET for Emulating Astrocyte and Dendrite Dynamics in Neuromorphic Systems

Neuromorphic systems seek to replicate the functionalities of biological neural networks to attain significant improvements in performance and efficiency of AI computing platforms. However, these systems have generally remained limited to emulation of simple neurons and synapses; and ignored higher order functionalities enabled by other components of the brain like astrocytes and dendrites. In this work, drawing inspiration from biology, we introduce a compact Double-Gate Ferroelectric Field Effect Transistor (DG-FeFET) cell that can emulate the dynamics of both astrocytes and dendrites within neuromorphic architectures. We demonstrate that with a ferroelectric top gate for synaptic weight programming as in conventional synapses and a non-ferroelectric back gate, the DG-FeFET realizes a synapse with a dynamic gain modulation mechanism. This can be leveraged as an analog for a compact astrocyte-tripartite synapse, as well as enabling dendrite-like gain modulation operations. By employing a fully-depleted silicon-on-insulator (FDSOI) FeFET as our double-gate device, we validate the linear control of the synaptic weight via the back gate terminal (i.e., the gate underneath the buried oxide (BOX) layer) through comprehensive theoretical and experimental studies. We showcase the promise such a tripartite synaptic device holds for numerous important neuromorphic applications, including autonomous self-repair of faulty neuromorphic hardware mediated by astrocytic functionality. Coordinate transformations based on dragonfly prey-interception circuitry models are also demonstrated based on dendritic function emulation by the device. This work paves the way forward for developing truly "brain-like" neuromorphic hardware that go beyond the current dogma focusing only on neurons and synapses.

cs.ET↗

arXiv2025

An In-Situ Spatial-Temporal Sequence Detector for Neuromorphic Vision Sensor Empowered by High Density Vertical NAND Storage

Neuromorphic vision sensors require efficient real-time pattern recognition, yet conventional architectures struggle with energy and latency constraints. Here, we present a novel in-situ spatiotemporal sequence detector that leverages vertical NAND storage to achieve massively parallel pattern detection. By encoding each cell with two single-transistor-based multi-level cell (MLC) memory elements, such as ferroelectric field-effect transistors (FeFETs), and mapping a pixel's temporal sequence onto consecutive word lines (WLs), we enable direct temporal pattern detection within NAND strings. Each NAND string serves as a dedicated reference for a single pixel, while different blocks store patterns for distinct pixels, allowing large-scale spatial-temporal pattern recognition via simple direct bit-line (BL) sensing, a well-established operation in vertical NAND storage. We experimentally validate our approach at both the cell and array levels, demonstrating that vertical NAND-based detector achieves more than six orders of magnitude improvement in energy efficiency and more than three orders of magnitude reduction in latency compared to conventional CPU-based methods. These findings establish vertical NAND storage as a scalable and energy-efficient solution for next-generation neuromorphic vision processing.

cs.ET↗