Lecture 1 Introduction
- Why should we care about deep multi-task & meta-learning?
deep learning allows us to handle unstructured inputs(pixels, languages, sensor readings, etc.)
without hand-engineering features, with less domain knowledge
Large, diverse data ——>deep learning——>broad generalization(能够实现广泛的推广)
- What if you need to quickly learn something new?
about a new person, for a new task, about a new environment, etc.
few-shot learning(少镜头学习)
by leveraging prior experience(利用先验知识)
This is where elements of multi-task learning can come into play.
- What is a task?
- Infromal problem definitions