In this paper, we introduced MIPSC: a novel and extensible selective classification model that effectively utilizes uncertainty in deep learning and combines it with predictive mean to make optimal decisions. We demonstrated MIPSC’s effectiveness ...
Numerous demonstrations of lexicase selection’s search performance have concluded it is superior to tournament selection on a variety of tasks. Previous attempts to explain this behavior have noted observations of increased population diversity, g...
We presented DE-DDQN, a Deep-RL-based operator selection method that learns to select online the mutation strategies of DE. DE-DDQN has two phases, offline training and online evaluation phase. During training we collected data from DE runs using ...
In this work we propose extending database systems with a persistent buffer pool on NVM. We consider such extension to impose a low implementation effort, since NVM is treated very similarly to DRAM during runtime, while leveraging well-understood...
Credentials & highlights
- Project Views