Constraining the Parameters of High-Dimensional Models with Active Learning. (arXiv:1905.08628v1 [cs.LG])
<a href="http://arxiv.org/find/cs/1/au:+Caron_S/0/1/0/all/0/1">Sascha Caron</a>, <a href="http://arxiv.org/find/cs/1/au:+Heskes_T/0/1/0/all/0/1">Tom Heskes</a>, <a href="http://arxiv.org/find/cs/1/au:+Otten_S/0/1/0/all/0/1">Sydney Otten</a>, <a href="http://arxiv.org/find/cs/1/au:+Stienen_B/0/1/0/all/0/1">Bob Stienen</a>
Constraining the parameters of physical models with $>5-10$ parameters is a
widespread problem in fields like particle physics and astronomy. In this paper
we show that this problem can be alleviated by the use of active learning. We
illustrate this with examples from high energy physics, a field where
computationally expensive simulations and large parameter spaces are common. We
show that the active learning techniques query-by-committee and
query-by-dropout-committee allow for the identification of model points in
interesting regions of high-dimensional parameter spaces (e.g. around decision
boundaries). This makes it possible to constrain model parameters more
efficiently than is currently done with the most common sampling algorithms.
Code implementing active learning can be found on GitHub.
Constraining the parameters of physical models with $>5-10$ parameters is a
widespread problem in fields like particle physics and astronomy. In this paper
we show that this problem can be alleviated by the use of active learning. We
illustrate this with examples from high energy physics, a field where
computationally expensive simulations and large parameter spaces are common. We
show that the active learning techniques query-by-committee and
query-by-dropout-committee allow for the identification of model points in
interesting regions of high-dimensional parameter spaces (e.g. around decision
boundaries). This makes it possible to constrain model parameters more
efficiently than is currently done with the most common sampling algorithms.
Code implementing active learning can be found on GitHub.
http://arxiv.org/icons/sfx.gif
