MAESTRO: A Data-Centric Approach to Understand Reuse, Performance, and Hardware Cost of DNN Mappings von Kwon, Hyoukjun, Chatarasi, Prasanth, Sarkar, Vivek, Krishna, Tushar, Pellauer, Michael, Parashar, Angshuman

Volltext

Evaluating Spatial Accelerator Architectures with Tiled Matrix-Matrix Multiplication von Moon, Gordon Euhyun, Kwon, Hyoukjun, Jeong, Geonhwa, Chatarasi, Prasanth, Rajamanickam, Sivasankaran, Krishna, Tushar

Volltext

A Communication-Centric Approach for Designing Flexible DNN Accelerators von Kwon, Hyoukjun, Samajdar, Ananda, Krishna, Tushar

Volltext

Evaluating Spatial Accelerator Architectures with Tiled Matrix-Matrix Multiplication von Moon, Gordon Euhyun, Kwon, Hyoukjun, Jeong, Geonhwa, Chatarasi, Prasanth, Rajamanickam, Sivasankaran, Krishna, Tushar

Volltext

Flexion: A Quantitative Metric for Flexibility in DNN Accelerators von Kwon, Hyoukjun, Pellauer, Michael, Parashar, Angshuman, Krishna, Tushar

Volltext

Architecture, Chip, and Package Codesign Flow for Interposer-Based 2.5-D Chiplet Integration Enabling Heterogeneous IP Reuse von Kim, Jinwoo, Murali, Gauthaman, Park, Heechun, Qin, Eric, Kwon, Hyoukjun, Chekuri, Venkata Chaitanya Krishna, Rahman, Nael Mizanur, Dasari, Nihar, Singh, Arvind, Lee, Minah, Torun, Hakki Mert, Roy, Kallol, Swaminathan, Madhavan, Mukhopadhyay, Saibal, Krishna, Tushar, Lim, Sung Kyu

Volltext

Marvel: A Data-Centric Approach for Mapping Deep Learning Operators on Spatial Accelerators von Chatarasi, Prasanth, Kwon, Hyoukjun, Parashar, Angshuman, Pellauer, Michael, Krishna, Tushar, Sarkar, Vivek

Volltext

A Formalism of DNN Accelerator Flexibility von Kao, Sheng-Chun, Kwon, Hyoukjun, Pellauer, Michael, Parashar, Angshuman, Krishna, Tushar

Volltext

Efficient Depth Estimation for Unstable Stereo Camera Systems on AR Glasses von Liu, Yongfan, Kwon, Hyoukjun

Volltext bestellen

Efficient Depth Estimation for Unstable Stereo Camera Systems on AR Glasses von Liu, Yongfan, Kwon, Hyoukjun

Volltext

FlexiBit: Fully Flexible Precision Bit-parallel Accelerator Architecture for Arbitrary Mixed Precision AI von Tahmasebi, Faraz, Wang, Yian, Huang, Benji Y. H, Kwon, Hyoukjun

Volltext bestellen

Performance Implications of Multi-Chiplet Neural Processing Units on Autonomous Driving Perception von Odema, Mohanad, Chen, Luke, Kwon, Hyoukjun, Faruque, Mohammad Abdullah Al

Volltext bestellen

Characterizing the Accuracy -- Efficiency Trade-off of Low-rank Decomposition in Language Models von Moar, Chakshu, Tahmasebi, Faraz, Pellauer, Michael, Kwon, Hyoukjun

Volltext bestellen

SCAR: Scheduling Multi-Model AI Workloads on Heterogeneous Multi-Chiplet Module Accelerators von Odema, Mohanad, Chen, Luke, Kwon, Hyoukjun, Faruque, Mohammad Abdullah Al

Volltext bestellen

NonGEMM Bench: Understanding the Performance Horizon of the Latest ML Workloads with NonGEMM Workloads von Karami, Rachid, Moar, Chakshu, Kao, Sheng-Chun, Kwon, Hyoukjun

Volltext bestellen

Inter-Layer Scheduling Space Exploration for Multi-model Inference on Heterogeneous Chiplets von Odema, Mohanad, Kwon, Hyoukjun, Faruque, Mohammad Abdullah Al

Volltext bestellen

A Formalism of DNN Accelerator Flexibility von Kao, Sheng-Chun, Kwon, Hyoukjun, Pellauer, Michael, Parashar, Angshuman, Krishna, Tushar

Volltext

Optimized Spatial Architecture Mapping Flow for Transformer Accelerators von Xu, Haocheng, Tahmasebi, Faraz, Qiao, Ye, Tian, Hongzheng, Kwon, Hyoukjun, Huang, Sitao

Volltext bestellen

PipeOrgan: Efficient Inter-operation Pipelining with Flexible Spatial Organization and Interconnects von Garg, Raveesh, Kwon, Hyoukjun, Qin, Eric, Chen, Yu-Hsin, Krishna, Tushar, Lai, Liangzhen

Volltext bestellen

DREAM: A Dynamic Scheduler for Dynamic Real-time Multi-model ML Workloads von Kim, Seah, Kwon, Hyoukjun, Song, Jinook, Jo, Jihyuck, Chen, Yu-Hsin, Lai, Liangzhen, Chandra, Vikas

Volltext bestellen