-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
421 lines (398 loc) · 18 KB
/
Copy pathindex.html
File metadata and controls
421 lines (398 loc) · 18 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
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
<!DOCTYPE HTML>
<html lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>Rico Zhu</title>
<meta name="author" content="Rico Zhu">
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="stylesheet" type="text/css" href="stylesheet.css">
</head>
<body>
<table
style="width:100%;max-width:800px;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr style="padding:0px">
<td style="padding:0px">
<table
style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr style="padding:0px">
<td style="padding:2.5%;width:63%;vertical-align:middle">
<p class="name" style="text-align: center;">
Rico Zhu
</p>
<p>I am a member of the technical staff at <a href="https://x.ai/">xAI</a>, where I work on the pre-training team for Grok. Before that I worked at <a href="https://tenstorrent.com/en">Tenstorrent</a>, under the AI models team.
I graduated in the spring of 2025 with highest distinction from Duke University, where I double majored in Computer Science and Math.
At Duke, I worked with Prof. <a href="https://users.cs.duke.edu/~cynthia/">Cynthia Rudin</a> on
interpretable machine learning research, with a focus on generative models.
</p>
<p>
I am fortunate to have also worked with Prof. <a href="https://sites.google.com/view/simonmak/home">Simon Mak</a>
as part of the <a href="https://jetscape.org/"> JETSCAPE </a> high-energy physics (HEP) collaboration,
Prof. <a href="https://www.yue-jiang.com/"> Yue Jiang</a> as part of the Interpretable ML Lab,
and Prof. <a href="https://users.cs.duke.edu/~rongge/">Rong Ge</a> on the theory of how transformers learn. My given name is Yifan, but I have
gone by Rico since about when I was 10, and is my preferred name.
</p>
<p style="text-align:center">
<a href="mailto:rico.zhu@duke.edu">Email</a> /
<a href="data/RicoZhu-CV.pdf">CV</a> /
<a href="https://scholar.google.com/citations?user=4WHneGEAAAAJ&hl=en">Google Scholar</a> /
<a href="https://github.com/ricohasgithub/">Github</a>
</p>
</td>
<td style="padding:2.5%;width:40%;max-width:40%">
<img
style="width:100%;max-width:100%;object-fit: cover; border-radius: 5%;" alt="profile photo"
src="images/Rico Zhu.jpg" class="hoverZoomLink">
</td>
</tr>
</tbody>
</table>
<table
style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:20px;width:100%;vertical-align:middle">
<h2>Research Interests</h2>
<p>
My interests broadly span applications of interpretable ML for science discovery, with an
emphasis in the high-energy physics (HEP) domain. I am interested in using symmetry to
design more interpretable and more realistic generative models for scientific emulation.
I worked on accelerating deep learning inference at Tenstorrent, and I currently work video generation pre/mid-training at xAI.
</p>
</td>
</tr>
</tbody>
</table>
<table>
<tr>
<td style="padding-left:20px;padding-top:20px;width:100%;vertical-align:middle">
<h2>Publications</h2>
</td>
</tr>
</table>
<table
style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/autoscha.png' style="padding-top:20px" width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="">
<span class="papertitle"> AutoSchA: Automatic Hierarchical Music Representations via Multi-Relational Node Isolation
</span>
</a>
<br>
Stephen Ni-Hahn*,
<strong>Rico Zhu*</strong>,
Jerry Yin,
Yue Jiang,
Cynthia Rudin,
Simon Mak
<br>
<em>AAAI, 2026</em>
<br>
<p></p>
<p>
Hierarchical representations provide powerful and principled approaches for analyzing many musical genres. We present a novel formulation of analysis using invariances, and design a graph neural network to complete this task.
</p>
</td>
</tr>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/climate_diffusion.png' style="padding-top:20px" width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://arxiv.org/pdf/2510.23866">
<span class="papertitle"> A PDE-Informed Latent Diffusion Model for 2-m Temperature Downscaling
</span>
</a>
<br>
Paul Rosu*,
Muchang Bahng*,
Erick Jiang*,
<strong>Rico Zhu*</strong>,
Vahid Tarokh
<br>
<em>Arxiv preprint, 2025</em>
<br>
<p></p>
<p>
We fine-tune a latent diffusion model for climate modelling, using a PDE-based auxilliary loss
to steer the model to produce more physically consistent images.
</p>
</td>
</tr>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/senthymnent.png' style="padding-top:20px" width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://dl.acm.org/doi/10.1145/3637528.3671626">
<span class="papertitle">SentHYMNent: An Interpretable and Sentiment-Driven Model for Algorithmic Melody Harmonization
</span>
</a>
<br>
Stephen Hahn,
Jerry Yin,
<strong>Rico Zhu</strong>,
Weihan Xu,
Yue Jiang,
Simon Mak,
Cynthia Rudin
<br>
<em>KDD</em>, 2024
<br>
<p></p>
<p>
Current music harmonization models fail to compose with <i>affect</i>, and those which do are uninterpretable.
We propose a novel affective embedding and sentiment representation in tandem with an efficient, interpretable generative model.
</p>
</td>
</tr>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/ismir_2024.png' width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://arxiv.org/pdf/2408.07184">
<span class="papertitle">A New Dataset, Notation Software, and Representation for Computational Schenkerian Analysis
</span>
</a>
<br>
Stephen Ni-Hahn,
Weihan Xu,
Jerry Yin,
<strong>Rico Zhu</strong>,
Simon Mak,
Yue Jiang,
Cynthia Rudin
<br>
<em>ISMIR</em>, 2024
<br>
<p></p>
<p>
Music theory is well understood under the framework of <i>Schenkerian Analysis</i>, a hierarchical
approach to harmonic analysis. We propose a novel graph-based formulation of this task.
</p>
</td>
</tr>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/schenkcomposer.png' width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://dl.acm.org/doi/abs/10.1145/3580305.3599772">
<span class="papertitle">An Interpretable, Flexible, and Interactive Probabilistic Framework for Melody Generation</span>
</a>
<br>
Stephen Hahn,
<strong>Rico Zhu</strong>,
Simon Mak,
Cynthia Rudin,
Yue Jiang
<br>
<em>KDD</em>, 2023
<br>
<p></p>
<p>
A novel generative framework for music composition that is architecturally compatible
with music theory, making the generation process easily interpretable for music experts.
</p>
</td>
</tr>
</tbody>
</table>
<table>
<tr>
<td style="padding-left:20px;padding-top:20px;width:100%;vertical-align:middle">
<h2>Demos</h2>
</td>
</tr>
</table>
<table
style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/nola_demo.png' style="padding-top:25px" width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://www.youtube.com/watch?v=AUqps10--U8">
<span class="papertitle">New Orleans: An Adventure in Music
</span>
</a>
<br>
Stephen Hahn,
<strong>Rico Zhu</strong>,
Jerry Yin,
Simon Mak,
Yue Jiang,
Cynthia Rudin
<br>
<em>NeurIPS Creative AI Track</em>, 2023
<br>
<p></p>
<p>
How can we generate music <i>affectively</i>? In this demo set in the city of New Orleans,
we present a music generation model which uses mixtures to compose with emotion.
</p>
</td>
</tr>
</tbody>
</table>
<table>
<tr>
<td style="padding-left:20px;padding-top:20px;width:100%;vertical-align:middle">
<h2>Research Experience</h2>
</td>
</tr>
</table>
<table
style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/icl1_layer0 1.gif' style="padding-top:20px" width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://jetscape.org/">
<span class="papertitle">Theory of In-Context Learning</span>
</a>
<br>
With Prof. Rong Ge.
<br>
May 2024 - May 2025
<br>
<p></p>
<p>
Theoretical analysis on the emergence of in-context and in-weights learning in a two-layer transformer network; analysis includes in-depth derivations of the model training dynamics in a synthetic setting, as well as conjectures and proofs for the optimal model weights.
</p>
</td>
</tr>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/jetscape.jpg' style="padding-top:20px" width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://jetscape.org/">
<span class="papertitle">JETSCAPE Collaboration</span>
</a>
<br>
Joint collaboration with Lawrence Berkeley National Lab, with Prof. Simon Mak
<br>
January 2024 - May 2025
<br>
<p></p>
<p>
Developed a Lorentz-invariant diffusion model for emulating particle collision events.
Designed a general, modular software framework for performing closure tests to verify experiment results.
</p>
</td>
</tr>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/duke_hep.jpg' style="padding-top:20px" width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://users.cs.duke.edu/~cynthia/lab.html">
<span class="papertitle">Interpretable Machine Learning Lab
</span>
</a>
<br>
With Prof. Cynthia Rudin, Simon Mak, and Yue Jiang
<br>
December 2022 - May 2025
<br>
<p></p>
<p>
Interpretable generative models for music composition, with a focus on designing models
with built-in interpretability under the Schenkerian music theory framework.
</p>
</td>
</tr>
<tr>
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<img src='images/cern.jpg' style="padding-top:25px" width=100%>
</div>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://ath11.pages.oit.duke.edu/flowaii/">
<span class="papertitle">CERN, Student Research Intern</span>
</a>
<br>
With the Duke HEP Group
<br>
May 2022 - September 2022
<br>
<p></p>
<a href="https://ath11.pages.oit.duke.edu/flowaii/">Documentation Page</a>
/
<a href="https://gitlab.oit.duke.edu/ath11/flowaii">Gitlab Repository</a>
<p>
With the ATLAS Collaboration, worked on using GNNs to model collider geometry for jet reconstruction as an alternative to the costly Particle Flow algorithm.
</p>
</td>
</tr>
</tbody>
</table>
<table width="100%" align="center" border="0" cellspacing="0" cellpadding="20">
<tbody>
<tr>
<td>
<h2>Teaching Experience</h2>
</td>
</tr>
</tbody>
</table>
<table width="100%" align="center" border="0" cellpadding="20">
<tbody>
<tr>
<td width="75%" valign="center">
Head Teaching Assistant, CS 201 (Data Structures & Algorithms) Spring 2023 - Spring 2025.
<br>
Teaching Assistant, CS 201 Fall 2022 - Spring 2025; CS 330 (Design & Analysis of Algorithms) Spring 2024 - Spring 2025.
<br>
Grader, Math 465 (High Dimensional Data Analysis) Fall 2024; Math 466 (Math of Machine Learning) Spring 2025.
</td>
</tr>
</tbody>
</table>
<table
style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
<tbody>
<tr>
<td style="padding:0px">
<br>
<p style="text-align:right;font-size:small;">
Souce code and website design from <a href="https://github.com/jonbarron/jonbarron_website">Jon Barron</a>.
</p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</table>
</body>
</html>