<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">This is an announcement of Han Liu's MS Presentation</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">===============================================</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">Candidate: Han Liu</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">Date: Friday, March 10, 2023</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
</div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">Time: 1:30 pm CST</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
</div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
<br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">Location: Searle 236</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">M.S. Paper Title: Learning Human-Compatible Representations for Case-Based Decision Support</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">Abstract: Algorithmic case-based decision support provides examples to help human make sense of predicted labels and aid human in decision-making
tasks. Despite the promising performance of supervised learning, representations learned by supervised models may not align well with human intuitions: what models consider as similar examples can be perceived as distinct by humans. As a result, they have
limited effectiveness in case-based decision support. In this work, we incorporate ideas from metric learning with supervised learning to examine the importance of alignment for effective decision support. In addition to instance-level labels, we use human-provided
triplet judgments to learn human-compatible decision-focused representations. Using both synthetic data and human subject experiments in multiple classification tasks, we demonstrate that such representation is better aligned with human perception than representation
solely optimized for classification. Human-compatible representations identify nearest neighbors that are perceived as more similar by humans and allow humans to make more accurate predictions, leading to substantial improvements in human decision accuracies
(17.8% in butterfly vs. moth classification and 13.2% in pneumonia classification).</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">Advisors: Chenhao Tan</span><br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<br style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto" class="ContentPasted0">
<span style="font-family:Helvetica;font-size:12px;orphans:auto;widows:auto;display:inline !important" class="ContentPasted0">Committee Members: Yuxin Chen, Chenhao Tan, Aritrick Chatterjee</span><br>
</div>
<div class="elementToProof">
<div id="Signature">
<div>
<div style="font-family: Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);">
<br>
</div>
</div>
</div>
</div>
</body>
</html>