Ginex: SSD-enabled Billion-scale Graph Neural Network Training on a Single Machine via Provably Optimal In-memory Caching.
Yeonhong ParkSunhong MinJae W. LeePublished in: CoRR (2022)
Keyphrases
- single machine
- neural network training
- scheduling problem
- processing times
- neural network
- total weighted tardiness
- earliness tardiness
- maximum lateness
- training algorithm
- release times
- minimize total
- release dates
- weighted number of tardy jobs
- sequence dependent setup times
- single machine scheduling problem
- scheduling jobs
- dynamic programming
- deteriorating jobs
- graph model
- optimization method
- search algorithm
- machine learning