Angshuman Parashar

Nvidia (United Kingdom) — United Kingdom

Computer Science · Physical Sciences

20h-index3.4kcitations62works3.52yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Parallel Computing and Optimization Techniques(33), Advanced Neural Network Applications(19), Embedded Systems Design Techniques(12), Interconnection Networks and Systems(11), Advanced Data Storage Technologies(10)

Publications62 total

SquareLoop: Explore Optimal Authentication Block Strategy for ML
2025Open Access
Mind the Gap: Attainable Data Movement and Operational Intensity Bounds for Tensor Algorithms
2024· 7 cited
HighLight: Efficient and Flexible DNN Acceleration with Hierarchical Structured Sparsity
2023· 22 citedOpen Access
LoopTree: Enabling Exploration of Fused-layer Dataflow Accelerators
2023· 7 cited
Demystifying Map Space Exploration for NPUs
2022· 7 cited
Demystifying Map Space Exploration for NPUs
arXiv (Cornell University)·2022Open Access
Sparseloop: An Analytical Approach To Sparse Tensor Accelerator Modeling
2022· 61 cited
A Formalism of DNN Accelerator Flexibility
ACM SIGMETRICS Performance Evaluation Review·2022· 1 cited
A Formalism of DNN Accelerator Flexibility
arXiv (Cornell University)·2022Open Access
A Formalism of DNN Accelerator Flexibility
2022· 2 cited
A Formalism of DNN Accelerator Flexibility
Proceedings of the ACM on Measurement and Analysis of Computing Systems·2022· 8 citedOpen Access
Sparseloop: An Analytical Approach To Sparse Tensor Accelerator Modeling
arXiv (Cornell University)·2022· 2 citedOpen Access
Ruby: Improving Hardware Efficiency for Tensor Algebra Accelerators Through Imperfect Factorization
2022· 13 cited
DiGamma: Domain-aware Genetic Algorithm for HW-Mapping Co-optimization for DNN Accelerators
2022 Design, Automation & Test in Europe Conference & Exhibition (DATE)·2022· 23 cited
DiGamma: Domain-aware Genetic Algorithm for HW-Mapping Co-optimization for DNN Accelerators
arXiv (Cornell University)·2022Open Access
Enabling Flexibility for Sparse Tensor Acceleration via Heterogeneity
arXiv (Cornell University)·2022· 3 citedOpen Access
Marvel: A Data-Centric Approach for Mapping Deep Learning Operators on Spatial Accelerators
ACM Transactions on Architecture and Code Optimization·2021· 30 citedOpen Access
Union: A Unified HW-SW Co-Design Ecosystem in MLIR for Evaluating Tensor Operations on Spatial Accelerators
arXiv (Cornell University)·2021Open Access
Union: A Unified HW-SW Co-Design Ecosystem in MLIR for Evaluating Tensor Operations on Spatial Accelerators
2021· 13 citedOpen Access
Union: A Unified HW-SW Co-Design Ecosystem in MLIR for Evaluating Tensor Operations on Spatial Accelerators.
2021· 1 citedOpen Access
page 1 of 4Next →

Frequent Co-authors

Michael Pellauer(28), Tushar Krishna(23), Joel Emer(20), Hyoukjun Kwon(17), Po-An Tsai(10), Prasanth Chatarasi(8), Michael Adler(8), Vivek Sarkar(7), Anand Sivasubramaniam(6), Sheng-Chun Kao(6), Yannan Nellie Wu(5), Vivienne Sze(5), Ananda Samajdar(5), Anurag Mukkara(4), Rangharajan Venkatesan(4), Brucek Khailany(4), Bushra Ahsan(4), Neal Crago(4), Daniel Lustig(4), В. В. Павлов(4)