Achieves lower convex hull of (N/(K-1), (K/(K-1)) * sum_{i=0 to M-1} (K/N)^i) pairs for K=2..N in secure private matrix multiplication over N servers.
Straggler Mitigation in Distributed Matrix Multiplication: Fundamental Limits and Optimal Coding
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2019 3verdicts
UNVERDICTED 3representative citing papers
Presents a coded distributed black-box optimization framework resilient to stragglers via error-correcting codes on search directions, extending evolution strategies, with experiments showing faster runtimes on adversarial attacks.
RKRP codes are MDS with probability 1, have identical communication/encoding costs to prior codes, lower average decoding complexity than OrthoPoly, and show substantially lower reconstruction error in numerical tests.
citing papers explorer
-
On the Upload versus Download Cost for Secure and Private Matrix Multiplication
Achieves lower convex hull of (N/(K-1), (K/(K-1)) * sum_{i=0 to M-1} (K/N)^i) pairs for K=2..N in secure private matrix multiplication over N servers.
-
Distributed Black-Box Optimization via Error Correcting Codes
Presents a coded distributed black-box optimization framework resilient to stragglers via error-correcting codes on search directions, extending evolution strategies, with experiments showing faster runtimes on adversarial attacks.
-
Random Khatri-Rao-Product Codes for Numerically-Stable Distributed Matrix Multiplication
RKRP codes are MDS with probability 1, have identical communication/encoding costs to prior codes, lower average decoding complexity than OrthoPoly, and show substantially lower reconstruction error in numerical tests.