Qirong Ho

This paper discusses continual learning of causal relations without losing previous knowledge.

A novel method for efficient hyperparameter tuning and cost management in AI training.