-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathcv.html
More file actions
276 lines (256 loc) · 19.7 KB
/
Copy pathcv.html
File metadata and controls
276 lines (256 loc) · 19.7 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Curriculum Vitae — David Xia</title>
<meta name="description" content="Curriculum Vitae of David Xia — theoretical machine learning and reinforcement learning.">
<link rel="icon" href="favicon.png">
<link rel="stylesheet" href="style.css">
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.5.0/css/all.min.css">
<script>document.documentElement.classList.add('js');(function(){try{var t=localStorage.getItem('theme');if(t==='dark'||(!t&&window.matchMedia('(prefers-color-scheme: dark)').matches)){document.documentElement.setAttribute('data-theme','dark');}}catch(e){}})();</script>
<script src="site.js" defer></script>
</head>
<body>
<nav class="nav">
<div class="nav-inner">
<a href="index.html" class="nav-brand">David Xia</a>
<div class="nav-links">
<a href="index.html">Home</a>
<a href="papers.html">Publications</a>
<a href="cv.html" class="active">CV</a>
<a href="pictures.html" class="nav-hide-sm">Gallery</a>
<a href="misc.html" class="nav-hide-sm">Misc</a>
<button class="theme-toggle" onclick="toggleTheme()" aria-label="Toggle dark mode" title="Toggle theme">
<i class="fas fa-moon"></i><i class="fas fa-sun"></i>
</button>
</div>
</div>
</nav>
<main class="wrap">
<header class="page-head">
<h1>Curriculum Vitae</h1>
<p>Theoretical machine learning & reinforcement learning · University of Illinois Urbana-Champaign</p>
<div class="cv-actions">
<a class="pill pill-primary" href="docs/david_xia_cv.pdf" target="_blank"><i class="fas fa-file-lines"></i> Open PDF</a>
<a class="pill" href="mailto:davidx3@illinois.edu"><i class="fas fa-envelope"></i> davidx3@illinois.edu</a>
</div>
</header>
<!-- ===== Research Interests ===== -->
<section class="cv-section">
<h2>Research Interests</h2>
<p>Theoretical ML, reinforcement learning, online learning, MDP function approximation, sample complexity.</p>
</section>
<!-- ===== Education ===== -->
<section class="cv-section">
<h2>Education</h2>
<div class="cv-entry">
<img class="cv-entry-logo" src="images/logos/uiuc_logo.png" alt="">
<div class="cv-entry-head">
<span class="cv-entry-title">University of Illinois Urbana-Champaign</span>
<span class="cv-entry-date">Aug 2023 – May 2027</span>
</div>
<div class="cv-entry-sub">Champaign, Illinois · GPA 3.99 / 4.00</div>
<ul class="cv-degrees">
<li>B.S. in Computer Science</li>
<li>B.S. in Liberal Arts & Sciences, Mathematics (Data Optimization concentration)</li>
<li>B.S. in Liberal Arts & Sciences, Statistics</li>
</ul>
</div>
</section>
<!-- ===== Papers ===== -->
<section class="cv-section">
<h2>Papers</h2>
<div class="pub-list">
<article class="pub">
<span class="pub-year">2025</span>
<div>
<div class="pub-title">RSK linear operators and the Vershik–Kerov–Logan–Shepp curve</div>
<div class="pub-authors"><span class="name">Duy Phan</span>, <span class="me">David Xia</span></div>
<div class="pub-venue"><span class="badge">TO APPEAR</span> Electronic Journal of Combinatorics</div>
<div class="link-row"><a class="chip arxiv" href="https://arxiv.org/abs/2504.05510" target="_blank"><i class="fas fa-file-lines"></i> arXiv</a></div>
</div>
</article>
<article class="pub">
<span class="pub-year">2025</span>
<div>
<div class="pub-title">EveGuard: Defeating Vibration-based Side-Channel Eavesdropping with Audio Adversarial Perturbations</div>
<div class="pub-authors"><span class="name">Jung-Woo Chang</span>, <span class="name">Ke Sun</span>, <span class="me">David Xia</span>, <span class="name">Xinyu Zhang</span>, <span class="name">Farinaz Koushanfar</span></div>
<div class="pub-venue"><span class="badge">PUBLISHED</span> IEEE Symposium on Security and Privacy 2025</div>
<div class="link-row"><a class="chip arxiv" href="https://arxiv.org/abs/2411.10034" target="_blank"><i class="fas fa-file-lines"></i> arXiv</a></div>
</div>
</article>
</div>
</section>
<!-- ===== Workshop Papers, Manuscripts, Notes ===== -->
<section class="cv-section">
<h2>Workshop Papers, Manuscripts & Notes</h2>
<div class="pub-list">
<article class="pub">
<span class="pub-year">2026</span>
<div>
<div class="pub-title">Frozen Policy Iteration for MDPs with Stochastic Transitions under Linear Q<sup>π</sup> Realizability</div>
<div class="pub-authors"><span class="me">David Xia</span>, <span class="name">Ruizhong Qiu</span>, <span class="name">Hanghang Tong</span></div>
<div class="pub-venue"><span class="badge">IN PREP</span> Manuscript in preparation</div>
<div class="link-row"><a class="chip" href="docs/stochastic_fpi.pdf" target="_blank"><i class="fas fa-file-pdf"></i> PDF</a></div>
</div>
</article>
<article class="pub">
<span class="pub-year">2026</span>
<div>
<div class="pub-title">Predictability in Major League Sports: Betting Odds versus Mathematical Models</div>
<div class="pub-authors"><span class="me">David Xia</span>, <span class="name">AJ Hildebrand</span></div>
<div class="pub-venue"><span class="badge">SUBMITTED</span> Mathematics and Sports (MAS) Journal</div>
<div class="link-row"><a class="chip" href="docs/predictability_paper.pdf" target="_blank"><i class="fas fa-file-pdf"></i> PDF</a></div>
</div>
</article>
<article class="pub">
<span class="pub-year">2026</span>
<div>
<div class="pub-title">Revitalizing Local Democracy: A Human-Centered Audit of LLMs in City Council Journalism</div>
<div class="pub-authors"><span class="me">David Xia</span>, <span class="name">Chris Maury</span></div>
<div class="pub-venue"><span class="badge">WORKSHOP</span> HEAL Workshop at CHI 2026</div>
<div class="link-row"><a class="chip" href="docs/heal_paper.pdf" target="_blank"><i class="fas fa-file-pdf"></i> PDF</a></div>
</div>
</article>
</div>
</section>
<!-- ===== Research Experience ===== -->
<section class="cv-section">
<h2>Research Experience</h2>
<div class="cv-entry">
<img class="cv-entry-logo" src="images/logos/northwestern_logo.png" alt="">
<div class="cv-entry-head">
<span class="cv-entry-title"><a class="name" href="https://zhaoranwang.github.io/" target="_blank">Northwestern University</a></span>
<span class="cv-entry-date">May 2026 – present</span>
</div>
<div class="cv-entry-sub">Research Intern · advised by <a class="name" href="https://zhaoranwang.github.io/" target="_blank">Prof. Zhaoran Wang</a></div>
<ul>
<li>Developed a closed-loop LLM meta-orchestrator agent for aircraft wing inverse design, where the agent diagnoses its own surrogate failures across iterative versions and emits corrective source code for the next optimization layer.</li>
<li>Identified and systematically addressed seven structural failure modes of surrogate-based optimization — including surrogate exploitation, Gaussian trust collapse, family-specific exploitation, and single-objective specification gaming — building a unified taxonomy and corresponding defenses.</li>
<li>Applied reinforcement-learning-style reasoning to the outer loop: the LLM agent classifies failure modes and upgrades its decision layer, achieving relative robustness gains validated against public RANS datasets (AirfRANS, SuperWing, ONERA CRM).</li>
</ul>
</div>
<div class="cv-entry">
<img class="cv-entry-logo" src="images/logos/isail_logo.png" alt="">
<div class="cv-entry-head">
<span class="cv-entry-title"><a class="name" href="https://www.idea-isail.com/index.html" target="_blank">iDEA-iSAIL Lab</a></span>
<span class="cv-entry-date">Feb 2026 – present</span>
</div>
<div class="cv-entry-sub">Research Intern · advised by <a class="name" href="http://tonghanghang.org/" target="_blank">Prof. Hanghang Tong</a></div>
<ul>
<li>Conducted research in theoretical reinforcement learning, studying sample and computation efficient algorithms for online learning in finite-horizon Markov Decision Processes under function approximation and stochastic dynamics.</li>
<li>Used concentrability to develop the first sample and computation efficient algorithm for online reinforcement learning in stochastic-transition MDPs under the linear Q<sup>π</sup> realizability assumption, resolving an open problem left by prior work restricted to deterministic dynamics.</li>
<li>Established a PAC sample complexity guarantee of Õ(d<sup>2</sup>H<sup>7</sup>C<sup>*</sup>/ε<sup>3</sup>) by combining techniques from offline and online reinforcement learning theory, dynamic programming, concentration analysis, and linear function approximation.</li>
<li>Manuscript in preparation: <a href="docs/stochastic_fpi.pdf" target="_blank">Frozen Policy Iteration for MDPs with Stochastic Transitions under Linear Q<sup>π</sup> Realizability</a>.</li>
</ul>
</div>
<div class="cv-entry">
<img class="cv-entry-logo" src="images/logos/iclue_logo.png" alt="">
<div class="cv-entry-head">
<span class="cv-entry-title"><a class="name" href="https://ayong.web.illinois.edu/ICLUE.html" target="_blank">Illinois Combinatorics Lab for Undergraduate Experience (ICLUE)</a></span>
<span class="cv-entry-date">Aug 2024 – present</span>
</div>
<div class="cv-entry-sub">Research Intern · advised by <a class="name" href="https://ayong.web.illinois.edu/" target="_blank">Prof. Alexander Yong</a></div>
<ul>
<li>Proved a new asymptotic result that the probability that the Schensted insertion algorithm for a uniformly random permutation in S<sub>n</sub> exhibits a special "bumping" interaction converges to 1 as n approaches infinity.</li>
<li>Derived the result using rigorous probability bounds on the RSK linear operator by leveraging the Vershik-Kerov-Logan-Shepp limit shape curve as the central analytic tool.</li>
<li>Developed proofs spanning real analysis (uniform convergence, limit arguments), probability theory (concentration inequalities, convergence in probability), and asymptotic analysis (bounding combinatorial quantities in the large-n regime).</li>
<li>Paper to appear in the <a href="https://www.combinatorics.org/" target="_blank">Electronic Journal of Combinatorics</a>.</li>
</ul>
</div>
<div class="cv-entry">
<img class="cv-entry-logo" src="images/logos/iml_logo.png" alt="">
<div class="cv-entry-head">
<span class="cv-entry-title"><a class="name" href="https://iml.math.illinois.edu/" target="_blank">Illinois Mathematics Lab</a></span>
<span class="cv-entry-date">Jan 2024 – May 2026</span>
</div>
<div class="cv-entry-sub">Research Intern · advised by <span class="name">Prof. AJ Hildebrand</span></div>
<ul>
<li>Researched the data-science perspective of sports matches, analyzing the predictability of outcomes by comparing prediction accuracies across polls, the betting market, Elo ratings, and mathematical models like Bradley-Terry.</li>
<li>Presented findings at the Rose-Hulman Undergraduate Math Conference 2024, UIUC Undergraduate Research Symposium 2024, and Joint Mathematics Meetings 2025.</li>
<li>Paper in submission to the <a href="https://janeway.uncpress.org/ms/" target="_blank">Mathematics and Sports (MAS) Journal</a>.</li>
</ul>
</div>
<div class="cv-entry">
<img class="cv-entry-logo" src="images/logos/cmu_logo.png" alt="">
<div class="cv-entry-head">
<span class="cv-entry-title"><a class="name" href="https://www.cs.cmu.edu/~jbigham/" target="_blank">Carnegie Mellon University, HCII</a></span>
<span class="cv-entry-date">May 2025 – Feb 2026</span>
</div>
<div class="cv-entry-sub">Research Intern · advised by <a class="name" href="https://www.cs.cmu.edu/~jbigham/" target="_blank">Prof. Jeff Bigham</a></div>
<ul>
<li>Designed and evaluated an automated journalism pipeline using SOTA LLMs to replicate end-to-end editorial workflows, from transcript segmentation to headline generation and topic prioritization.</li>
<li>Orchestrated a large-scale crowd-sourced study and demonstrated that LLM-based headline quality and topic prioritization can meet and exceed professional standards.</li>
<li>Paper in the CHI 2026 <a href="https://heal-workshop.github.io/" target="_blank">HEAL workshop</a>.</li>
</ul>
</div>
<div class="cv-entry">
<img class="cv-entry-logo" src="images/logos/nus_logo.png" alt="">
<div class="cv-entry-head">
<span class="cv-entry-title"><a class="name" href="https://cde.nus.edu.sg/ece/staff/fong-xuanyao-kelvin/" target="_blank">National University of Singapore, SERIUS REU</a></span>
<span class="cv-entry-date">May 2024 – Aug 2024</span>
</div>
<div class="cv-entry-sub">Research Intern · advised by <a class="name" href="https://cde.nus.edu.sg/ece/staff/fong-xuanyao-kelvin/" target="_blank">Prof. Kelvin Fong Xuanyao</a></div>
<ul>
<li>Designed the GARDEN (Generalized Anomaly Recognition and Detection for Enhanced Nurturing) pipeline for early disease detection in farm plants, combining segmentation, generative, and classification models end-to-end.</li>
<li>Built a leaf segmentation model to isolate regions of interest from complex backgrounds, enabling downstream anomaly detection to focus on plant tissue only.</li>
<li>Implemented a conditional GAN (cGAN) to synthesize realistic diseased and healthy leaf images, augmenting training data and improving classifier generalization.</li>
<li>Achieved robust disease classification by ensembling deep learning and traditional ML models, exploiting complementary strengths to improve detection reliability on real farm images.</li>
</ul>
</div>
<div class="cv-entry">
<img class="cv-entry-logo" src="images/logos/ucsd_logo.png" alt="">
<div class="cv-entry-head">
<span class="cv-entry-title"><a class="name" href="https://xyzhang.ucsd.edu/index.html" target="_blank">University of California San Diego</a></span>
<span class="cv-entry-date">Jun 2023 – Dec 2024</span>
</div>
<div class="cv-entry-sub">Research Intern · advised by <a class="name" href="https://xyzhang.ucsd.edu/index.html" target="_blank">Prof. Xinyu Zhang</a></div>
<ul>
<li>Developed EveGuard, a software-based defense framework to protect voice privacy from vibrometry-based side channels, using a perturbation generator model (PGM) to suppress sensor-based eavesdropping while preserving high audio quality.</li>
<li>Implemented Eve-GAN, a novel domain-translation task for inferring eavesdropped signals, enabling end-to-end training of the PGM and using few-shot learning to reduce data-collection overhead.</li>
<li>Achieved a protection rate of over 97% against audio classifiers, hindering eavesdropped reconstruction.</li>
<li>Paper published in the <a href="https://sp2025.ieee-security.org/" target="_blank">IEEE Symposium on Security and Privacy 2025</a>.</li>
</ul>
</div>
</section>
<!-- ===== Grants & Awards ===== -->
<section class="cv-section">
<h2>Grants & Awards</h2>
<ul class="award-list">
<li><span class="yr">2025</span><span>CMU HCII REU — NSF 2349558</span></li>
<li><span class="yr">2025</span><span>ICLUE RTG — NSF 1937241</span></li>
</ul>
</section>
<!-- ===== Coursework ===== -->
<section class="cv-section">
<h2>Relevant Coursework</h2>
<div class="cv-coursework">
<p><span class="cat">CS:</span> Statistical Reinforcement Learning, Machine Learning, NLP, Deep Learning, Algorithms, Numerical Methods, Numerical Analysis</p>
<p><span class="cat">Math:</span> Honors Real Analysis, Graph Theory, Combinatorics, Abstract Linear Algebra, Abstract Algebra I–II, Algebraic Combinatorics, Linear Programming</p>
<p><span class="cat">Stats:</span> Stochastic Processes, Applied Bayesian Analysis, Survival Analysis</p>
</div>
</section>
<!-- ===== Languages ===== -->
<section class="cv-section">
<h2>Languages</h2>
<div class="cv-coursework">
<p><span class="cat">English:</span> Native</p>
<p><span class="cat">Mandarin Chinese:</span> Fluent, HSK6</p>
</div>
</section>
</main>
<footer class="footer">
<div class="wrap footer-inner">
<div class="footer-social">
<a href="mailto:davidx3@illinois.edu" aria-label="Email"><i class="fas fa-envelope"></i></a>
<a href="https://scholar.google.com/citations?hl=en&user=uScYa6cAAAAJ" target="_blank" aria-label="Google Scholar"><i class="fas fa-graduation-cap"></i></a>
<a href="https://github.com/davidxia3" target="_blank" aria-label="GitHub"><i class="fab fa-github"></i></a>
<a href="https://www.linkedin.com/in/dx3/" target="_blank" aria-label="LinkedIn"><i class="fab fa-linkedin-in"></i></a>
</div>
<span class="leet" title="David Xia">|)/\/||) ><|/\</span>
<span class="footer-meta">© 2026 David Xia</span>
</div>
</footer>
</body>
</html>