arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.07464cs.LGmath.OCstat.ML
收藏

探索的超模型

Hypermodels for Exploration

DeepMind(深度思维)

Vikranth Dwaracherla, Xiuyuan Lu, Morteza Ibrahimi, Ian Osband, Zheng Wen, Benjamin Van Roy

摘要:

我们研究使用超模型来表示认知的不确定性和指导探索。这推广并扩展了合奏的使用,以近似汤普森采样。训练集成的计算成本随着其大小而增长,因此,先前的工作通常限于具有数十个元素的集成。我们发现,替代超模型可以获得显着的效率提升,实现否则需要数百或数千个元素的行为,甚至在集成方法无法学习的情况下取得成功。这允许更准确的近似汤普森采样以及使用更复杂的勘探计划。特别是,我们认为一个近似形式的信息导向的采样,并证明相对于汤普森采样的性能增益。作为集成的替代方案,我们考虑线性和神经网络超模型,也称为超网络。我们证明,与神经网络的基础模型,一个线性超模型可以表示基本上任何分布的功能,因此,超网络是没有更多的表达。

英文摘要:

We study the use of hypermodels to represent epistemic uncertainty and guide exploration. This generalizes and extends the use of ensembles to approximate Thompson sampling. The computational cost of training an ensemble grows with its size, and as such, prior work has typically been limited to ensembles with tens of elements. We show that alternative hypermodels can enjoy dramatic efficiency gains, enabling behavior that would otherwise require hundreds or thousands of elements, and even succeed in situations where ensemble methods fail to learn regardless of size. This allows more accurate approximation of Thompson sampling as well as use of more sophisticated exploration schemes. In particular, we consider an approximate form of information-directed sampling and demonstrate performance gains relative to Thompson sampling. As alternatives to ensembles, we consider linear and neural network hypermodels, also known as hypernetworks. We prove that, with neural network base models, a linear hypermodel can represent essentially any distribution over functions, and as such, hypernetworks are no more expressive.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑