Search Results for author:
Found 25 papers, 4 papers with code
Ev-Edge: Efficient Execution of Event-based Vision Algorithms on Commodity Edge Platforms
On several state-of-art networks for a range of autonomous navigation tasks, Ev-Edge achieves 1. 28x-2. 05x improvements in latency and 1. 23x-2. 15x in energy over an all-GPU implementation on the NVIDIA Jetson Xavier AGX platform for single-task execution scenarios.
Input Compression with Positional Consistency for Efficient Training and Inference of Transformer Neural Networks
This leads to smaller input sequences being processed by the Transformer, and hence faster training, while also alleviating overfitting by presenting each input with different compression levels.
Evaluation of STT-MRAM as a Scratchpad for Training in ML Accelerators
We devised a cross-layer simulation framework to evaluate the effectiveness of STT-MRAM as a scratchpad replacing SRAM in a systolic-array-based DNN accelerator.
X-Former: In-Memory Acceleration of Transformers
Transformers have achieved great success in a wide variety of natural language processing (NLP) tasks due to the attention mechanism, which assigns an importance score for every word relative to other words in a sequence.
Approximate Computing and the Efficient Machine Learning Expedition
In this work, we enlighten the synergistic nature of AxC and ML and elucidate the impact of AxC in designing efficient ML systems.
A Co-design view of Compute in-Memory with Non-Volatile Elements for Neural Networks
In the second part, we review what is knows about the different new non-volatile memory materials and devices suited for compute in-memory, and discuss the outlook and challenges.
Efficient Ensembles of Graph Neural Networks
In this work, we leverage the unique characteristics of GNNs to overcome these overheads, creating efficient ensemble GNNs that are faster than even single models at inference time.
Specialized Transformers: Faster, Smaller and more Accurate NLP Models
We propose a Specialization framework to create optimized transformer models for a given downstream task.
InterTrain: Accelerating DNN Training using Input Interpolation
This is because the efficacy of learning on interpolated inputs is reduced by the interference between the forward/backward propagation of their constituent inputs.
PIM-DRAM: Accelerating Machine Learning Workloads using Processing in Commodity DRAM
Processing in memory has been proposed as a promising solution for the memory wall bottleneck for ML workloads.
Accelerating DNN Training through Selective Localized Learning
The trend in the weight updates made to the transition layer across epochs is used to determine how the boundary betweenSGD and localized updates is shifted in future epochs.
Ax-BxP: Approximate Blocked Computation for Precision-Reconfigurable Deep Neural Network Acceleration
To efficiently support precision re-configurability in DNN accelerators, we introduce an approximate computing method wherein DNN computations are performed block-wise (a block is a group of bits) and re-configurability is supported at the granularity of blocks.
AxFormer: Accuracy-driven Approximation of Transformers for Faster, Smaller and more Accurate NLP Models
We propose AxFormer, a systematic framework that applies accuracy-driven approximations to create optimized transformer models for a given downstream task.
Sparsity Turns Adversarial: Energy and Latency Attacks on Deep Neural Networks
We also evaluate the impact of the attack on a sparsity-optimized DNN accelerator and demonstrate degradations up to 1. 59x in latency, and also study the performance of the attack on a sparsity-optimized general-purpose processor.
EMPIR: Ensembles of Mixed Precision Deep Networks for Increased Robustness against Adversarial Attacks
Our results indicate that EMPIR boosts the average adversarial accuracies by 42. 6%, 15. 2% and 10. 5% for the DNN models trained on the MNIST, CIFAR-10 and ImageNet datasets respectively, when compared to single full-precision models, without sacrificing accuracy on the unperturbed inputs.
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.
TxSim:Modeling Training of Deep Neural Networks on Resistive Crossbar Systems
To address this challenge, there is a need for tools that can model the functional impact of non-idealities on DNN training and inference.
Gradual Channel Pruning while Training using Feature Relevance Scores for Convolutional Neural Networks
To address all the above issues, we present a simple-yet-effective gradual channel pruning while training methodology using a novel data-driven metric referred to as feature relevance score.
TiM-DNN: Ternary in-Memory accelerator for Deep Neural Networks
The use of lower precision has emerged as a popular technique to optimize the compute and storage requirements of complex Deep Neural Networks (DNNs).
Pack and Detect: Fast Object Detection in Videos Using Region-of-Interest Packing
Based on these observations, we propose Pack and Detect (PaD), an approach to reduce the computational requirements of object detection in videos.
RxNN: A Framework for Evaluating Deep Neural Networks on Resistive Crossbars
We present RxNN, a fast and accurate simulation framework to evaluate large-scale DNNs on resistive crossbar systems.
SparCE: Sparsity aware General Purpose Core Extensions to Accelerate Deep Neural Networks
SparCE consists of 2 key micro-architectural enhancements- a Sparsity Register File (SpRF) that tracks zero registers and a Sparsity aware Skip Address (SASA) table that indicates instructions to be skipped.
DyVEDeep: Dynamic Variable Effort Deep Neural Networks
Complementary to these approaches, DyVEDeep is a dynamic approach that exploits the heterogeneity in the inputs to DNNs to improve their compute efficiency with comparable classification accuracy.
Multiplier-less Artificial Neurons Exploiting Error Resiliency for Energy-Efficient Neural Computing
Multipliers consume most of the processing energy in the digital neurons, and thereby in the hardware implementations of artificial neural networks.
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.
