98 citations · 270 across the 20 of their papers we have counts for
21 papers
Restoring the Broken Covenant Between Compilers and Deep Learning Accelerators
Sean Kinzer, Soroush Ghodrati, Rohan Mahapatra +6
Deep learning accelerators address the computational demands of Deep Neural Networks (DNNs), departing from the traditional Von Neumann execution model. They leverage specialized h…
An Open-Source ML-Based Full-Stack Optimization Framework for Machine Learning Accelerators
Hadi Esmaeilzadeh, Soroush Ghodrati, Andrew B. Kahng +8
Parameterizable machine learning (ML) accelerators are the product of recent breakthroughs in ML. To fully enable their design space exploration (DSE), we propose a physical-design…
Performance Analysis of DNN Inference/Training with Convolution and non-Convolution Operations
Hadi Esmaeilzadeh, Soroush Ghodrati, Andrew B. Kahng +4
Today's performance analysis frameworks for deep learning accelerators suffer from two significant limitations. First, although modern convolutional neural network (CNNs) consist o…
In-Storage Domain-Specific Acceleration for Serverless Computing
Rohan Mahapatra, Soroush Ghodrati, Byung Hoon Ahn +8
While (1) serverless computing is emerging as a popular form of cloud execution, datacenters are going through major changes: (2) storage dissaggregation in the system infrastructu…
Accelerating Attention through Gradient-Based Learned Runtime Pruning
Zheng Li, Soroush Ghodrati, Amir Yazdanbakhsh +2
Self-attention is a key enabler of state-of-art accuracy for various transformer-based Natural Language Processing models. This attention mechanism calculates a correlation score f…
Conscious AI
Hadi Esmaeilzadeh, Reza Vaezi
Recent advances in artificial intelligence (AI) have achieved human-scale speed and accuracy for classification tasks. In turn, these capabilities have made AI a viable replacement…
WaveQ: Gradient-Based Deep Quantization of Neural Networks through Sinusoidal Adaptive Regularization
Ahmed T. Elthakeb, Prannoy Pilligundla, Fatemehsadat Mireshghallah +3
As deep neural networks make their ways into different domains, their compute efficiency is becoming a first-order constraint. Deep quantization, which reduces the bitwidth of the…
Bit-Parallel Vector Composability for Neural Acceleration
Soroush Ghodrati, Hardik Sharma, Cliff Young +2
Conventional neural accelerators rely on isolated self-sufficient functional units that perform an atomic operation while communicating the results through an operand delivery-aggr…
Privacy in Deep Learning: A Survey
Fatemehsadat Mireshghallah, Mohammadkazem Taram, Praneeth Vepakomma +3
The ever-growing advances of deep learning in many areas including vision, recommendation systems, natural language processing, etc., have led to the adoption of Deep Neural Networ…
Ordering Chaos: Memory-Aware Scheduling of Irregularly Wired Neural Networks for Edge Devices
Byung Hoon Ahn, Jinwon Lee, Jamie Menjay Lin +3
Recent advances demonstrate that irregularly wired neural networks from Neural Architecture Search (NAS) and Random Wiring can not only automate the design of deep neural networks…
Not All Features Are Equal: Discovering Essential Features for Preserving Prediction Privacy
Fatemehsadat Mireshghallah, Mohammadkazem Taram, Ali Jalali +3
When receiving machine learning services from the cloud, the provider does not need to receive all features; in fact, only a subset of the features are necessary for the target pre…
Chameleon: Adaptive Code Optimization for Expedited Deep Neural Network Compilation
Byung Hoon Ahn, Prannoy Pilligundla, Amir Yazdanbakhsh +1
Achieving faster execution with shorter compilation time can foster further diversity and innovation in neural networks. However, the current paradigm of executing neural networks…
Mixed-Signal Charge-Domain Acceleration of Deep Neural networks through Interleaved Bit-Partitioned Arithmetic
Soroush Ghodrati, Hardik Sharma, Sean Kinzer +5
Low-power potential of mixed-signal design makes it an alluring option to accelerate Deep Neural Networks (DNNs). However, mixed-signal circuitry suffers from limited range for inf…
Divide and Conquer: Leveraging Intermediate Feature Representations for Quantized Training of Neural Networks
Ahmed T. Elthakeb, Prannoy Pilligundla, Alex Cloninger +1
The deep layers of modern neural networks extract a rather rich set of features as an input propagates through the network. This paper sets out to harvest these rich intermediate r…
Reinforcement Learning and Adaptive Sampling for Optimized DNN Compilation
Byung Hoon Ahn, Prannoy Pilligundla, Hadi Esmaeilzadeh
Achieving faster execution with shorter compilation time can enable further diversity and innovation in neural networks. However, the current paradigm of executing neural networks…
Shredder: Learning Noise Distributions to Protect Inference Privacy
Fatemehsadat Mireshghallah, Mohammadkazem Taram, Prakash Ramrakhyani +2
A wide variety of deep neural applications increasingly rely on the cloud to perform their compute-heavy inference. This common practice requires sending private and privileged dat…
SinReQ: Generalized Sinusoidal Regularization for Low-Bitwidth Deep Quantized Training
Ahmed T. Elthakeb, Prannoy Pilligundla, Hadi Esmaeilzadeh
Deep quantization of neural networks (below eight bits) offers significant promise in reducing their compute and storage cost. Albeit alluring, without special techniques for train…
ReLeQ: A Reinforcement Learning Approach for Deep Quantization of Neural Networks
Ahmed T. Elthakeb, Prannoy Pilligundla, FatemehSadat Mireshghallah +2
Deep Neural Networks (DNNs) typically require massive amount of computation resource in inference tasks for computer vision applications. Quantization can significantly reduce DNN…
GANAX: A Unified MIMD-SIMD Acceleration for Generative Adversarial Networks
Amir Yazdanbakhsh, Hajar Falahati, Philip J. Wolfe +3
Generative Adversarial Networks (GANs) are one of the most recent deep learning models that generate synthetic data from limited genuine datasets. GANs are on the frontier as furth…
In-RDBMS Hardware Acceleration of Advanced Analytics
Divya Mahajan, Joon Kyung Kim, Jacob Sacks +3
The data revolution is fueled by advances in machine learning, databases, and hardware design. Programmable accelerators are making their way into each of these areas independently…
Bit Fusion: Bit-Level Dynamically Composable Architecture for Accelerating Deep Neural Networks
Hardik Sharma, Jongse Park, Naveen Suda +5
Fully realizing the potential of acceleration for Deep Neural Networks (DNNs) requires understanding and leveraging algorithmic properties. This paper builds upon the algorithmic i…