<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
<span style="font-size: 14.666666984558105px;" class="">This is an announcement of Fengxue Zhang's MS Presentation</span><br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">===============================================</span><br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">Candidate: Fengxue Zhang</span><br style="font-size: 14.666666984558105px;" class="">
<br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">Date: Tuesday, April 11, 2023</span><br style="font-size: 14.666666984558105px;" class="">
<br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">Time: 1:00 pm CST</span><br style="font-size: 14.666666984558105px;" class="">
<br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">Location: JCL 298</span><br style="font-size: 14.666666984558105px;" class="">
<br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">M.S. Paper Title: Learning Regions of Interest for Bayesian Optimization with Adaptive Level-Set Estimation</span><br style="font-size: 14.666666984558105px;" class="">
<br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">Abstract: We study Bayesian optimization (BO) in high-dimensional and non-stationary scenarios. Existing algorithms for such scenarios typically require extensive hyperparameter tuning, which limits their
practical effectiveness. We propose a framework, called BALLET, which adaptively filters for a high-confidence region of interest (ROI) as a superlevel-set of a nonparametric probabilistic model such as a Gaussian process (GP). Our approach is easy to tune,
and is able to focus on local region of the optimization space that can be tackled by existing BO methods. The key idea is to use two probabilistic models: a coarse GP to identify the ROI, and a localized GP for optimization within the ROI. We show theoretically
that BALLET can efficiently shrink the search space, and can exhibit a tighter regret bound than standard BO without ROI filtering. We demonstrate empirically the effectiveness of BALLET on both synthetic and real-world optimization tasks.</span><br style="font-size: 14.666666984558105px;" class="">
<br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">Advisors: Yuxin Chen</span><br style="font-size: 14.666666984558105px;" class="">
<br style="font-size: 14.666666984558105px;" class="">
<span style="font-size: 14.666666984558105px;" class="">Committee Members: Rebecca Willett, Haifeng Xu, and Yuxin Chen</span><br class="">
<div class=""><br class="Apple-interchange-newline">
</div>
<br class="">
</body>
</html>