Stochastic Weight Averaging in Parallel: Large-Batch Training That Generalizes Well.
Vipul GuptaSantiago Akle SerranoDennis DeCostePublished in: ICLR (2020)
Keyphrases
- batch mode
- parallel processing
- test set
- training algorithm
- training examples
- multistage
- monte carlo
- information retrieval
- training phase
- training process
- stochastic programming
- parallel programming
- stochastic model
- parallel computing
- parallel implementation
- shared memory
- steady state
- back propagation
- object detection
- learning algorithm