Search Results for author:
Found 88 papers, 25 papers with code
MEADOW: Memory-efficient Dataflow and Data Packing for Low Power Edge LLMs
While prior LLM-targeted quantization, and prior works on sparse acceleration have significantly mitigated the memory and computation bottleneck, they do so assuming high power platforms such as GPUs and server-class FPGAs with large off-chip memory bandwidths and employ a generalized matrix multiplication (GEMM) execution of all the layers in the decoder.
Low-power Spike-based Wearable Analytics on RRAM Crossbars
This work introduces a spike-based wearable analytics system utilizing Spiking Neural Networks (SNNs) deployed on an In-memory Computing engine based on RRAM crossbars, which are known for their compactness and energy-efficiency.
Intelligent Sensing-to-Action for Robust Autonomy at the Edge: Opportunities and Challenges
Autonomous edge computing in robotics, smart cities, and autonomous vehicles relies on the seamless integration of sensing, processing, and actuation for real-time decision-making in dynamic environments.
TesseraQ: Ultra Low-Bit LLM Post-Training Quantization with Block Reconstruction
To effectively optimize the rounding in LLMs and stabilize the reconstruction process, we introduce progressive adaptive rounding.
Spiking Transformer with Spatial-Temporal Attention
Spike-based Transformer presents a compelling and energy-efficient alternative to traditional Artificial Neural Network (ANN)-based Transformers, achieving impressive results through sparse binary computations.
ReSpike: Residual Frames-based Hybrid Spiking Neural Networks for Efficient Action Recognition
Spiking Neural Networks (SNNs) have emerged as a compelling, energy-efficient alternative to traditional Artificial Neural Networks (ANNs) for static image tasks such as image classification and segmentation.
When In-memory Computing Meets Spiking Neural Networks -- A Perspective on Device-Circuit-System-and-Algorithm Co-design
This review explores the intersection of bio-plausible artificial intelligence in the form of Spiking Neural Networks (SNNs) with the analog In-Memory Computing (IMC) domain, highlighting their collective potential for low-power edge computing environments.
TReX- Reusing Vision Transformer's Attention for Efficient Xbar-based Computing
However, all prior works have neglected the overhead and co-depencence of attention blocks on the accuracy-energy-delay-area of IMC-implemented ViTs.
LoAS: Fully Temporal-Parallel Dataflow for Dual-Sparse Spiking Neural Networks
We observe that naively running a dual-sparse SNN on existing spMspM accelerators designed for dual-sparse Artificial Neural Networks (ANNs) exhibits sub-optimal efficiency.
One-stage Prompt-based Continual Learning
With the QR loss, our approach maintains ~ 50% computational cost reduction during inference as well as outperforms the prior two-stage PCL methods by ~1. 4% on public class-incremental continual learning benchmarks including CIFAR-100, ImageNet-R, and DomainNet.
ClipFormer: Key-Value Clipping of Transformers on Memristive Crossbars for Write Noise Mitigation
Specifically, we find pre-trained Vision Transformers (ViTs) to be vulnerable on crossbars due to the impact of write noise on the dynamically-generated Key (K) and Value (V) matrices in the attention layers, an effect not accounted for in prior studies.
TT-SNN: Tensor Train Decomposition for Efficient Spiking Neural Network Training
Spiking Neural Networks (SNNs) have gained significant attention as a potentially energy-efficient alternative for standard neural networks with their sparse binary activation.
GenQ: Quantization in Low Data Regimes with Generative Synthetic Data
In the realm of deep neural network deployment, low-bit quantization presents a promising avenue for enhancing computational efficiency.
Rethinking Skip Connections in Spiking Neural Networks with Time-To-First-Spike Coding
In this work, we delve into the role of skip connections, a widely used concept in Artificial Neural Networks (ANNs), within the domain of SNNs with TTFS coding.
Are SNNs Truly Energy-efficient? $-$ A Hardware Perspective
Spiking Neural Networks (SNNs) have gained attention for their energy-efficient machine learning capabilities, utilizing bio-inspired activation functions and sparse binary spike-data representations.
RobustEdge: Low Power Adversarial Detection for Cloud-Edge Systems
However, in prior detection works, the detector is attached to the classifier model and both detector and classifier work in tandem to perform adversarial detection that requires a high computational overhead which is not available at the low-power edge.
Artificial to Spiking Neural Networks Conversion for Scientific Machine Learning
We introduce a method to convert Physics-Informed Neural Networks (PINNs), commonly used in scientific machine learning, to Spiking Neural Networks (SNNs), which are expected to have higher energy efficiency compared to traditional Artificial Neural Networks (ANNs).
Examining the Role and Limits of Batchnorm Optimization to Mitigate Diverse Hardware-noise in In-memory Computing
In-Memory Computing (IMC) platforms such as analog crossbars are gaining focus as they facilitate the acceleration of low-precision Deep Neural Networks (DNNs) with high area- & compute-efficiencies.
Input-Aware Dynamic Timestep Spiking Neural Networks for Efficient In-Memory Computing
Although the efficiency of SNNs can be realized on the In-Memory Computing (IMC) architecture, we show that the energy cost and latency of SNNs scale linearly with the number of timesteps used on IMC hardware.
Do We Really Need a Large Number of Visual Prompts?
Due to increasing interest in adapting models on resource-constrained edges, parameter-efficient transfer learning has been widely explored.
Sharing Leaky-Integrate-and-Fire Neurons for Memory-Efficient Spiking Neural Networks
Spiking Neural Networks (SNNs) have gained increasing attention as energy-efficient neural networks owing to their binary and asynchronous computation.
MINT: Multiplier-less INTeger Quantization for Energy Efficient Spiking Neural Networks
We propose Multiplier-less INTeger (MINT) quantization, a uniform quantization scheme that efficiently compresses weights and membrane potentials in spiking neural networks (SNNs).
Divide-and-Conquer the NAS puzzle in Resource Constrained Federated Learning Systems
In this paper, we propose DC-NAS -- a divide-and-conquer approach that performs supernet-based Neural Architecture Search (NAS) in a federated system by systematically sampling the search space.
Uncovering the Representation of Spiking Neural Networks Trained with Surrogate Gradient
However, some essential questions exist pertaining to SNNs that are little studied: Do SNNs trained with surrogate gradient learn different representations from traditional Artificial Neural Networks (ANNs)?
NeuroBench: A Framework for Benchmarking Neuromorphic Computing Algorithms and Systems
To address these shortcomings, we present NeuroBench: a benchmark framework for neuromorphic computing algorithms and systems.
SEENN: Towards Temporal Spiking Early-Exit Neural Networks
However, we observe that the information capacity in SNNs is affected by the number of timesteps, leading to an accuracy-efficiency tradeoff.
XPert: Peripheral Circuit & Neural Architecture Co-search for Area and Energy-efficient Xbar-based Computing
The hardware-efficiency and accuracy of Deep Neural Networks (DNNs) implemented on In-memory Computing (IMC) architectures primarily depend on the DNN architecture and the peripheral circuit parameters.
XploreNAS: Explore Adversarially Robust & Hardware-efficient Neural Architectures for Non-ideal Xbars
This work proposes a two-phase algorithm-hardware co-optimization approach called XploreNAS that searches for hardware-efficient & adversarially robust neural architectures for non-ideal crossbar platforms.
Workload-Balanced Pruning for Sparse Spiking Neural Networks
Though the existing pruning methods can provide extremely high weight sparsity for deep SNNs, the high weight sparsity brings a workload imbalance problem.
DeepCAM: A Fully CAM-based Inference Accelerator with Variable Hash Lengths for Energy-efficient Deep Neural Networks
The first innovation is an approximate dot-product built on computations in the Euclidean space that can replace addition and multiplication with simple bit-wise operations.
Exploring Temporal Information Dynamics in Spiking Neural Networks
After training, we observe that information becomes highly concentrated in earlier few timesteps, a phenomenon we refer to as temporal information concentration.
Wearable-based Human Activity Recognition with Spatio-Temporal Spiking Neural Networks
SNNs allow spatio-temporal extraction of features and enjoy low-power computation with binary spikes.
SpikeSim: An end-to-end Compute-in-Memory Hardware Evaluation Tool for Benchmarking Spiking Neural Networks
To this end, we propose SpikeSim, a tool that can perform realistic performance, energy, latency and area evaluation of IMC-mapped SNNs.
Exploring Lottery Ticket Hypothesis in Spiking Neural Networks
To scale up a pruning technique towards deep SNNs, we investigate Lottery Ticket Hypothesis (LTH) which states that dense networks contain smaller subnetworks (i. e., winning tickets) that achieve comparable performance to the dense networks.
Examining the Robustness of Spiking Neural Networks on Non-ideal Memristive Crossbars
Spiking Neural Networks (SNNs) have recently emerged as the low-power alternative to Artificial Neural Networks (ANNs) owing to their asynchronous, sparse, and binary information processing.
MIME: Adapting a Single Neural Network for Multi-task Inference with Memory-efficient Dynamic Pruning
Recent years have seen a paradigm shift towards multi-task learning.
SATA: Sparsity-Aware Training Accelerator for Spiking Neural Networks
Based on SATA, we show quantitative analyses of the energy efficiency of SNN training and compare the training cost of SNNs and ANNs.
Addressing Client Drift in Federated Continual Learning with Adaptive Optimization
However, there is little attention towards additional challenges emerging when federated aggregation is performed in a continual learning system.
Neuromorphic Data Augmentation for Training Spiking Neural Networks
In an effort to minimize this generalization gap, we propose Neuromorphic Data Augmentation (NDA), a family of geometric augmentations specifically designed for event-based datasets with the goal of significantly stabilizing the SNN training and reducing the generalization gap between training and test performance.
Ranked #1 on
Event data classification
on CIFAR10-DVS
(using extra training data)
Adversarial Detection without Model Information
We train a standalone detector independent of the classifier model, with a layer-wise energy separation (LES) training to increase the separation between natural and adversarial energies.
Rate Coding or Direct Coding: Which One is Better for Accurate, Robust, and Energy-efficient Spiking Neural Networks?
Then, we measure the robustness of the coding techniques on two adversarial attack methods.
Neural Architecture Search for Spiking Neural Networks
Interestingly, SNASNet found by our search algorithm achieves higher performance with backward connections, demonstrating the importance of designing SNN architecture for suitably using temporal information.
Examining and Mitigating the Impact of Crossbar Non-idealities for Accurate Implementation of Sparse Deep Neural Networks
Although, these techniques have claimed to preserve the accuracy of the sparse DNNs on crossbars, none have studied the impact of the inexorable crossbar non-idealities on the actual performance of the pruned networks.
Gradient-based Bit Encoding Optimization for Noise-Robust Binary Memristive Crossbar
In addition, we propose Gradient-based Bit Encoding Optimization (GBO) which optimizes a different number of pulses at each layer, based on our in-depth analysis that each layer has a different level of noise sensitivity.
Beyond Classification: Directly Training Spiking Neural Networks for Semantic Segmentation
Specifically, we first investigate two representative SNN optimization techniques for recognition tasks (i. e., ANN-SNN conversion and surrogate gradient learning) on semantic segmentation datasets.
RAPID-RL: A Reconfigurable Architecture with Preemptive-Exits for Efficient Deep-Reinforcement Learning
To address this challenge, we propose a reconfigurable architecture with preemptive exits for efficient deep RL (RAPID-RL).
Federated Learning with Spiking Neural Networks
To validate the proposed federated learning framework, we experimentally evaluate the advantages of SNNs on various aspects of federated learning with CIFAR10 and CIFAR100 benchmarks.
Efficiency-driven Hardware Optimization for Adversarially Robust Neural Networks
In this paper, we show how the bit-errors in the 6T cells of hybrid 6T-8T memories minimize the adversarial perturbations in a DNN.
PrivateSNN: Privacy-Preserving Spiking Neural Networks
2) Class leakage is caused when class-related features can be reconstructed from network parameters.
Visual Explanations from Spiking Neural Networks using Interspike Intervals
Spiking Neural Networks (SNNs) compute and communicate with asynchronous binary temporal events that can lead to significant energy savings with neuromorphic hardware.
Activation Density based Mixed-Precision Quantization for Energy Efficient Neural Networks
Additionally, we find that integrating AD based quantization with AD based pruning (both conducted during training) yields up to ~198x and ~44x energy reductions for VGG19 and ResNet18 architectures respectively on PIM platform compared to baseline 16-bit precision, unpruned models.
Noise Sensitivity-Based Energy Efficient and Robust Adversary Detection in Neural Networks
Based on prior works on detecting adversaries, we propose a structured methodology of augmenting a deep neural network (DNN) with a detector subnetwork.
Exposing the Robustness and Vulnerability of Hybrid 8T-6T SRAM Memory Architectures to Adversarial Attacks in Deep Neural Networks
In this work, we elicit the advantages and vulnerabilities of hybrid 6T-8T memories to improve the adversarial robustness and cause adversarial attacks on DNNs.
Revisiting Batch Normalization for Training Low-latency Deep Spiking Neural Networks from Scratch
Different from previous works, our proposed BNTT decouples the parameters in a BNTT layer along the time axis to capture the temporal dynamics of spikes.
Compression-aware Continual Learning using Singular Value Decomposition
Specifically, we decompose the weight filters using SVD and train the network on incremental tasks in its factorized form.
Rethinking Non-idealities in Memristive Crossbars for Adversarial Robustness in Neural Networks
Deep Neural Networks (DNNs) have been shown to be prone to adversarial attacks.
Domain Adaptation without Source Data
Our key idea is to leverage a pre-trained model from the source domain and progressively update the target model in a self-learning manner.
Enabling Deep Spiking Neural Networks with Hybrid Conversion and Spike Timing Dependent Backpropagation
We propose a hybrid training methodology: 1) take a converted SNN and use its weights and thresholds as an initialization step for spike-based backpropagation, and 2) perform incremental spike-timing dependent backpropagation (STDB) on this carefully initialized network to obtain an SNN that converges within few epochs and requires fewer time steps for input processing.
QUANOS- Adversarial Noise Sensitivity Driven Hybrid Quantization of Neural Networks
We identify a novel noise stability metric (ANS) for DNNs, i. e., the sensitivity of each layer's computation to adversarial noise.
Inherent Adversarial Robustness of Deep Spiking Neural Networks: Effects of Discrete Input Encoding and Non-Linear Activations
Our results suggest that SNNs trained with LIF neurons and smaller number of timesteps are more robust than the ones with IF (Integrate-Fire) neurons and larger number of timesteps.
Pruning Filters while Training for Efficiently Optimizing Deep Learning Networks
Our results for VGG-16 trained on CIFAR10 shows that L1 normalization provides the best performance among all the techniques explored in this work with less than 1% drop in accuracy after pruning 80% of the filters compared to the original network.
Energy-efficient and Robust Cumulative Training with Net2Net Transformation
We achieve this by first training a small network (with lesser parameters) on a small subset of the original dataset, and then gradually expanding the network using Net2Net transformation to train incrementally on larger subsets of the dataset.
Relevant-features based Auxiliary Cells for Energy Efficient Detection of Natural Errors
To address this issue, we propose an ensemble of classifiers at hidden layers to enable energy efficient detection of natural errors.
Activation Density driven Energy-Efficient Pruning in Training
Neural network pruning with suitable retraining can yield networks with considerably fewer parameters than the original with comparable degrees of accuracy.
Towards Scalable, Efficient and Accurate Deep Spiking Neural Networks with Backward Residual Connections, Stochastic Softmax and Hybridization
Spiking Neural Networks (SNNs) may offer an energy-efficient alternative for implementing deep learning applications.
Synthesizing Images from Spatio-Temporal Representations using Spike-based Backpropagation
The spiking autoencoders are benchmarked on MNIST and Fashion-MNIST and achieve very low reconstruction loss, comparable to ANNs.
Evaluating the Stability of Recurrent Neural Models during Training with Eigenvalue Spectra Analysis
We analyze the stability of recurrent networks, specifically, reservoir computing models during training by evaluating the eigenvalue spectra of the reservoir dynamics.
A Comprehensive Analysis on Adversarial Robustness of Spiking Neural Networks
In this work, we present, for the first time, a comprehensive analysis of the behavior of more bio-plausible networks, namely Spiking Neural Network (SNN) under state-of-the-art adversarial tests.
Enabling Spike-based Backpropagation for Training Deep Neural Network Architectures
Spiking Neural Networks (SNNs) have recently emerged as a prominent neural computing paradigm.
Discretization based Solutions for Secure Machine Learning against Adversarial Attacks
Specifically, discretizing the input space (or allowed pixel levels from 256 values or 8-bit to 4 values or 2-bit) extensively improves the adversarial robustness of DLNs for a substantial range of perturbations for minimal loss in test accuracy.
A Low Effort Approach to Structured CNN Design Using PCA
We demonstrate the proposed methodology on AlexNet and VGG style networks on the CIFAR-10, CIFAR-100 and ImageNet datasets, and successfully achieve an optimized architecture with a reduction of up to 3. 8X and 9X in the number of operations and parameters respectively, while trading off less than 1% accuracy.
Implicit Generative Modeling of Random Noise during Training for Adversarial Robustness
We introduce a Noise-based prior Learning (NoL) approach for training neural networks that are intrinsically robust to adversarial attacks.
Exploiting Inherent Error-Resiliency of Neuromorphic Computing to achieve Extreme Energy-Efficiency through Mixed-Signal Neurons
In this work, we will analyze, compare and contrast existing neuron architectures with a proposed mixed-signal neuron (MS-N) in terms of performance, power and noise, thereby demonstrating the applicability of the proposed mixed-signal neuron for achieving extreme energy-efficiency in neuromorphic computing.
Tree-CNN: A Hierarchical Deep Convolutional Neural Network for Incremental Learning
Over the past decade, Deep Convolutional Neural Networks (DCNNs) have shown remarkable performance in most computer vision tasks.
Chaos-guided Input Structuring for Improved Learning in Recurrent Neural Networks
Anatomical studies demonstrate that brain reformats input information to generate reliable responses for performing computations.
An Energy-Efficient Mixed-Signal Neuron for Inherently Error-Resilient Neuromorphic Systems
This work presents the design and analysis of a mixed-signal neuron (MS-N) for convolutional neural networks (CNN) and compares its performance with a digital neuron (Dig-N) in terms of operating frequency, power and noise.
Learning to Recognize Actions from Limited Training Examples Using a Recurrent Spiking Neural Model
A fundamental challenge in machine learning today is to build a model that can learn from few examples.
STDP Based Pruning of Connections and Weight Quantization in Spiking Neural Networks for Energy Efficient Recognition
We present a sparse SNN topology where non-critical connections are pruned to reduce the network size and the remaining critical synapses are weight quantized to accommodate for limited conductance levels.
Voltage-Driven Domain-Wall Motion based Neuro-Synaptic Devices for Dynamic On-line Learning
The basic building blocks of such neuromorphic systems are neurons and synapses.
Gabor Filter Assisted Energy Efficient Fast Learning Convolutional Neural Networks
This combination creates a balanced system that gives better training performance in terms of energy and time, compared to the standalone CNN (without any Gabor kernels), in exchange for tolerable accuracy degradation.
ASP: Learning to Forget with Adaptive Synaptic Plasticity in Spiking Neural Networks
Against this backdrop, we present a novel unsupervised learning mechanism ASP (Adaptive Synaptic Plasticity) for improved recognition with Spiking Neural Networks (SNNs) for real time on-line learning in a dynamic environment.
Convolutional Spike Timing Dependent Plasticity based Feature Learning in Spiking Neural Networks
Brain-inspired learning models attempt to mimic the cortical architecture and computations performed in the neurons and synapses constituting the human brain to achieve its efficiency in cognitive tasks.
RESPARC: A Reconfigurable and Energy-Efficient Architecture with Memristive Crossbars for Deep Spiking Neural Networks
In this paper, we propose RESPARC - a reconfigurable and energy efficient architecture built-on Memristive Crossbar Arrays (MCA) for deep Spiking Neural Networks (SNNs).
FALCON: Feature Driven Selective Classification for Energy-Efficient Image Recognition
We evaluate our approach for a 12-object classification task on the Caltech101 dataset and 10-object task on CIFAR-10 dataset by constructing FALCON models on the NeuE platform in 45nm technology.
Attention Tree: Learning Hierarchies of Visual Features for Large-Scale Image Recognition
A set of binary classifiers is organized on top of the learnt hierarchy to minimize the overall test-time complexity.
Unsupervised Regenerative Learning of Hierarchical Features in Spiking Deep Networks for Object Recognition
We present a spike-based unsupervised regenerative learning scheme to train Spiking Deep Networks (SpikeCNN) for object recognition problems using biologically realistic leaky integrate-and-fire neurons.
Conditional Deep Learning for Energy-Efficient and Enhanced Pattern Recognition
Deep learning neural networks have emerged as one of the most powerful classification tools for vision related applications.
Energy-Efficient Object Detection using Semantic Decomposition
We propose a 2-stage hierarchical classification framework, with increasing levels of complexity, wherein the first stage is trained to recognize the broad representative semantic features relevant to the object of interest.
