This Project Grant from the National Science Foundation's $125,000 Computer and Information Science and Engineering program will fund the development of optimized sparse tensor network algorithms and specialized accelerator architectures for heterogeneous computing systems. Awarded to North Carolina State University on October 1, 2022 for a one-year period ending September 30, 2023, the grant supports preliminary research to address challenges in sparse tensor network computations through four approaches: memory heterogeneity-aware representations and data re-arrangement; balanced sparse tensor contraction algorithms with optimized data placement; memoization and intelligent resource allocation; and specialized accelerators. The optimized sparse tensor networks encompass high-performance computing, algorithms, compilers, computer architecture, and performance modeling efforts to be tested under multiple application scenarios.