It is shown that this instability can propagate across echelons and compound over time, even when the underlying demand path is held fixed, and two approaches for improving reliability are evaluated: reinforcement-learning post-training and operational guardrails.
C. Long, David Simchi-Levi, Feng Zhu et al.· arXiv.org· 2 citations
It is proved that such speedups in sample-complexity require only $\tilde O\left( 1 \right)$ communication, substantially improving upon the communication costs in prior work.
Sreejeet Maity, Feng Zhu, Aritra Mitra et al.· American Control Conference· 0 citations
This work proposes Batched SGD, a simple variant that partitions online samples into epochs and performs a single update per epoch using a refined, low-variance gradient estimate, and provides the first high-probability guarantees for FL.
Feng Zhu, Robert W. Heath, Aritra Mitra· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.