Algorithm design for tensor units

DOI10.1007/978-3-030-85665-6_22MaRDI QIDQ6487193zbMATH OpenFDO

Authors Rezaul Alam Chowdhury, Francesco Silvestri, Flavio Vella

Publication date 31 March 2022

Full work available at URL https://doi.org/10.1007/978-3-030-85665-6_22

Mathematics Subject Classification ID

Symbolic computation and algebraic computation (68W30) Analysis of algorithms (68W40) Hardware implementations of nonnumerical algorithms (VLSI algorithms, etc.) (68W35)

Recommendations

• Strassen's Algorithm for Tensor Contraction
• Design of a high-performance GEMM-like tensor-tensor multiplication
• Discovering faster matrix multiplication algorithms with reinforcement learning
• Convolution accelerator designs using fast algorithms
• Arithmetic unit design for neural accelerators: cost performance issues

Cites work

• Algorithms and data structures for external memory
• Fast Output-Sensitive Matrix Multiplication
• Gaussian elimination is not optimal
• Graph expansion and communication costs of fast matrix multiplication
• Introduction to algorithms
• On the Complexity of Matrix Product
• On the all-pairs-shortest-path problem in unweighted undirected graphs.
• Sorting and Permuting without Bank Conflicts on GPUs

Cited in

(2)

• Arithmetic unit design for neural accelerators: cost performance issues
• Discovering faster matrix multiplication algorithms with reinforcement learning

This page was built for publication: Algorithm design for tensor units

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6487193)