University of Cambridge > Talks.cam > Machine Learning Reading Group @ CUED > Meta-reinforcement learning

Meta-reinforcement learning

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Elre Oldewage.

Meta learning allows for generalisation across tasks and has become increasingly relevant as machine learning systems are asked to solve heterogeneous problems efficiently with less training data. In recent years, meta learning has been applied in the context of reinforcement learning to build agents that learn to generalise across a distribution of Markov decision problems. In this reading group, we will briefly introduce the basics of meta reinforcement learning, cover different approaches to the problem, and discuss their uses and limitations. We will also consider how they compare to more traditional algorithms, both learned and hand-crafted.

Recommended reading:

- Wang et al. 2016 (https://arxiv.org/abs/1611.05763) OR Duan et al. 2016 (https://arxiv.org/abs/1611.02779).

- Finn et al. 2017 (https://arxiv.org/abs/1703.03400). Nagabandi et al. 2019 (https://arxiv.org/abs/1803.11347).

- This blog post also provides an overview of several of the topics we will cover: https://lilianweng.github.io/lil-log/2019/06/23/meta-reinforcement-learning.html

This talk is part of the Machine Learning Reading Group @ CUED series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2021 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity