CS PhD Student at Brown | Computational Design, Geometry, AI
Joined May 2022
- Tweets 55
- Following 135
- Followers 178
- Likes 431
5 Photos and videos
15 Dec 2025
Can we apply gradient descent to discrete changes? In our new #SIGGRAPHAsia paper, we show that gradient descent can work on shape grammars, as in CAD and procedural modeling, but only if the grammars are designed correctly!
6
42
262
64,733
15 Dec 2025
We show that these properties matter through ablations over grammar variants and compare against existing random-walk approaches like RJMCMC. Grammars that satisfy more of our guidelines consistently converge faster and reach better optima.
1
1
18
1,281
15 Dec 2025
Our takeaway: Don’t just throw an optimizer at arbitrary grammars, design the grammar for descent!
Paper: Design for Descent: What Makes a Shape Grammar Easy to Optimize?
Authors: @milin_k_, @jackzzhang, @nmwsharp, @AdrianaSchulz7
Project: computationaldesign.group/pu…
2
1
25
1,175
Jack Zhang retweeted
31 Jul 2024
Jack was one of the first students in our group 3DL! He did a couple of awesome works with us during his undergrad. Check out his papers and make sure to follow his journey as he starts as a PhD student in @uwcse with @AdrianaSchulz7 ! zzhang-18.github.io/
1
1
12
1,021
Jack Zhang retweeted
3 Aug 2022
One thing that makes siggraph unique is that it’s so much more than just technical papers. I’m excited that @UChicagoCS will be active across several different forums: technical papers, workshops, emerging technologies, frontiers talks, and even a social meetup! Come meet us 🤩
2 Aug 2022
#SIGGRAPH2022 here we come! My lab is doing:
(1) #XR/#VR meetup (w/ @michinebeling ) (2) Yudai Tanaka's (@Hoku5ai) electrical head actuation demo, (3) #haptics talk w/ Nicholas Colonnese (@RealityLabs), (4) metaverse workshop w/@marknb00, @michinebeling & @tengshanyuan join us!
6
21
Jack Zhang retweeted
6 May 2022
Thrilled to share "GANimator: Neural Motion Synthesis from a Single Sequence", in this upcoming #SIGGRAPH2022 🤩
GANimator can produce novel animations for unique creatures that don't have large motion datasets! For example, this hexapedal crab 🦀(1/4)
8
56
320