Phantom 2D Accelerator

Phantom 2D accelerators are specialized hardware designed to efficiently process sparse convolutional neural networks (CNNs), aiming to improve the speed and energy efficiency of deep learning computations. Current research focuses on optimizing architectures for various CNN layers (including those with non-unit strides) and employing techniques like dynamic thread scheduling and multi-level load balancing to maximize hardware utilization. These advancements are significant because they address the limitations of existing accelerators in handling the irregular data structures inherent in sparse CNNs, leading to improved performance in applications ranging from image recognition to robotic control.

Papers

November 5, 2024

SpiDR: A Reconfigurable Digital Compute-in-Memory Spiking Neural Network Accelerator for Event-based Perception
Deepika Sharma, Shubham Negi, Trishit Dutta, Amogh Agrawal, Kaushik Roy
Spiking Neural Network Event Based Vision Computing in Memory Phantom 2D Accelerator

October 15, 2024

DPD-NeuralEngine: A 22-nm 6.6-TOPS/W/mm$^2$ Recurrent Neural Network Accelerator for Wideband Power Amplifier Digital Pre-Distortion
Ang Li, Haolin Wu, Yizhuo Wu, Qinyu Chen, Leo C. N. de Vreede, Chang Gao
Recurrent Neural Network Gated Recurrent Unit Phantom 2D Accelerator Neural CDEs Complementary Metal Oxide Semiconductor Digital Pre Distortion

October 8, 2024

Provable Accuracy Bounds for Hybrid Dynamical Optimization and Sampling
Matthew X. Burns, Qingyuan Hou, Michael C. Huang
Open Sampling Error Bound Phantom 2D Accelerator Analog Bit Hybrid Optimization Neighborhood Search Device Variation Analog Accelerator

September 17, 2024

FSL-HDnn: A 5.7 TOPS/W End-to-end Few-shot Learning Classifier Accelerator with Feature Extraction and Hyperdimensional Computing
Haichao Yang, Chang Eun Song, Weihong Xu, Behnam Khaleghi, Uday Mallappa, Monil Shah, Keming Fan, Mingu Kang, Tajana Rosing
Feature Extraction Shot Classification Hyperdimensional Computing Phantom 2D Accelerator

May 21, 2024

Adaptive Robotic Arm Control with a Spiking Recurrent Neural Network on a Digital Accelerator
Alejandro Linares-Barranco, Luciano Prono, Robert Lengenstein, Giacomo Indiveri, Charlotte Frenkel
Neuron Model Chip Design Phantom 2D Accelerator

April 13, 2024

Deep Reinforcement Learning based Online Scheduling Policy for Deep Neural Network Multi-Tenant Multi-Accelerator Systems
Francesco G. Blanco, Enrico Russo, Maurizio Palesi, Davide Patti, Giuseppe Ascia, Vincenzo Catania
Deep Neural Network Deep Reinforcement Learning Phantom 2D Accelerator Multi Accelerator Online Scheduling

January 28, 2024

Anomaly Detection of Particle Orbit in Accelerator using LSTM Deep Learning Technology
Zhiyuan Chen, Wei Lu, Radhika Bhong, Yimin Hu, Brian Freeman, Adam Carpenter
Anomaly Detection Orbital Motion Phantom 2D Accelerator Proton Beam

December 28, 2023

Beyond PID Controllers: PPO with Neuralized PID Policy for Proton Beam Intensity Control in Mu2e
Chenwei Xu, Jerry Yao-Chieh Hu, Aakaash Narayanan, Mattson Thieme, Vladimir Nagaslaev, Mark Austin, Jeremy Arnold, Jose Berlioz, Pierrick Hanlet, Aisha Ibrahim, Dennis Nicklaus, Jovan Mitrevski, Jason Michael St. John, Gauri Pradhan, Andrea Saewert, Kiyomi Seiya, Brian Schupbach, Randy Thurman-Keup, Nhan Tran, Rui Shi, Seda Ogrenci, Alexis Maya-Isabelle Shuping, Kyle Hazelwood, Han Liu
Proximal Policy Optimization Policy Optimization Phantom 2D Accelerator Proton Beam Proportional Integral Derivative

December 14, 2023

Inter-Layer Scheduling Space Exploration for Multi-model Inference on Heterogeneous Chiplets
Mohanad Odema, Hyoukjun Kwon, Mohammad Abdullah Al Faruque
Phantom 2D Accelerator Cross Layer Multi Model

November 16, 2023

Stella Nera: Achieving 161 TOp/s/W with Multiplier-free DNN Acceleration based on Approximate Matrix Multiplication
Jannis Schönleber, Lukas Cavigelli, Renzo Andri, Matteo Perotti, Luca Benini
Matrix Multiplication Phantom 2D Accelerator Differential Analysis Efficient Hardware Multiply Accumulate TOp Front TopMost

August 14, 2023

Symphony: Optimized DNN Model Serving using Deferred Batch Scheduling
Lequn Chen, Weixin Deng, Anirudh Canumalla, Yu Xin, Danyang Zhuo, Matthai Philipose, Arvind Krishnamurthy
DNN Framework DNN Model Dynamic Scheduling DNN Inference Phantom 2D Accelerator

July 5, 2023

Distance Preserving Machine Learning for Uncertainty Aware Accelerator Capacitance Predictions
Steven Goldenberg, Malachi Schram, Kishansingh Rajput, Thomas Britton, Chris Pappas, Dan Lu, Jared Walden, Majdi I. Radaideh, Sarah Cousineau, Sudarshan Harave
Gaussian Process Phantom 2D Accelerator Accurate Uncertainty Capacitive Sensing Distance Preserving

April 12, 2023

Exploiting Logic Locking for a Neural Trojan Attack on Machine Learning Accelerators
Hongye Xu, Dongfang Liu, Cory Merkel, Michael Zuzak
Logic Reasoning Phantom 2D Accelerator Machine Learning Accelerator Neural Backdoor Neural Trojan Trigger Attack

April 10, 2023

Deploying Machine Learning Models to Ahead-of-Time Runtime on Edge Using MicroTVM
Chen Liu, Matthias Jobst, Liyuan Guo, Xinyue Shi, Johannes Partzsch, Christian Mayr
Machine Learning Code Generation Extreme Edge Hand Gesture Recognition Runtime Analysis AI Application Phantom 2D Accelerator Intensive Operator

February 3, 2023

An Optical XNOR-Bitcount Based Accelerator for Efficient Inference of Binary Neural Networks
Sairam Sri Vatsavai, Venkata Sai Praneeth Karempudi, Ishan Thakkar
Efficient Inference Binary Neural Network Optical System Phantom 2D Accelerator BNN Layer XOR Gate

December 6, 2022

Integration of a systolic array based hardware accelerator into a DNN operator auto-tuning framework
F. N. Peccia, O. Bringmann
Hardware Accelerator Phantom 2D Accelerator Auto Tuning Systolic Array Heterogeneous System on Chip Custom Accelerator

November 4, 2022

An Efficient FPGA-based Accelerator for Deep Forest
Mingyu Zhu, Jiapeng Luo, Wendong Mao, Zhongfeng Wang
High Efficiency Field Programmable Gate Array Efficient Architecture Phantom 2D Accelerator Compute Node Deep Forest Deep Regression Forest

October 7, 2022

Multi-objective and multi-fidelity Bayesian optimization of laser-plasma acceleration
Faran Irshad, Stefan Karsch, Andreas Döpp
Multi Objective Optimization Multi Objective Bayesian Optimization Phantom 2D Accelerator Beam Prediction Multi Fidelity Bayesian Optimization

August 26, 2022

DiVa: An Accelerator for Differentially Private Machine Learning
Beomsik Park, Ranggi Hwang, Dongho Yoon, Yoonhyuk Choi, Minsoo Rhu
Differential Privacy Differential Privacy SGD Private Machine Learning Phantom 2D Accelerator Diva 360

June 28, 2022

H-GCN: A Graph Convolutional Network Accelerator on Versal ACAP Architecture
Chengming Zhang, Tong Geng, Anqi Guo, Jiannan Tian, Martin Herbordt, Ang Li, Dingwen Tao
Graph Neural Network Phantom 2D Accelerator GNN Inference GCN Accelerator

Phantom 2D Accelerator

Papers

SpiDR: A Reconfigurable Digital Compute-in-Memory Spiking Neural Network Accelerator for Event-based Perception

DPD-NeuralEngine: A 22-nm 6.6-TOPS/W/mm$^2$ Recurrent Neural Network Accelerator for Wideband Power Amplifier Digital Pre-Distortion

Provable Accuracy Bounds for Hybrid Dynamical Optimization and Sampling

FSL-HDnn: A 5.7 TOPS/W End-to-end Few-shot Learning Classifier Accelerator with Feature Extraction and Hyperdimensional Computing

Adaptive Robotic Arm Control with a Spiking Recurrent Neural Network on a Digital Accelerator

Deep Reinforcement Learning based Online Scheduling Policy for Deep Neural Network Multi-Tenant Multi-Accelerator Systems

Anomaly Detection of Particle Orbit in Accelerator using LSTM Deep Learning Technology

Beyond PID Controllers: PPO with Neuralized PID Policy for Proton Beam Intensity Control in Mu2e

Inter-Layer Scheduling Space Exploration for Multi-model Inference on Heterogeneous Chiplets

Stella Nera: Achieving 161 TOp/s/W with Multiplier-free DNN Acceleration based on Approximate Matrix Multiplication

Symphony: Optimized DNN Model Serving using Deferred Batch Scheduling

Distance Preserving Machine Learning for Uncertainty Aware Accelerator Capacitance Predictions

Exploiting Logic Locking for a Neural Trojan Attack on Machine Learning Accelerators

Deploying Machine Learning Models to Ahead-of-Time Runtime on Edge Using MicroTVM

An Optical XNOR-Bitcount Based Accelerator for Efficient Inference of Binary Neural Networks

Integration of a systolic array based hardware accelerator into a DNN operator auto-tuning framework

An Efficient FPGA-based Accelerator for Deep Forest

Multi-objective and multi-fidelity Bayesian optimization of laser-plasma acceleration

DiVa: An Accelerator for Differentially Private Machine Learning

H-GCN: A Graph Convolutional Network Accelerator on Versal ACAP Architecture