Pingchuan Ma

Ph.D. Candidate, MIT CSAIL

Email: pcma (at) csail.mit.edu

Office: MIT Stata Center 32-321

Biography

Pingchuan Ma (馬平川) is a fifth-year Ph.D. candidate at MIT CSAIL advised by Prof. Wojciech Matusik. Previously, he worked with Prof. Bo Ren and Prof. Ming-Ming Cheng at Nankai University. His research goal covers the whole pipeline of physics-based intelligence: (1) reconstructing efficient realistic physics-based environments, (2) generating intelligent agents upon the environments, and (3) realizing the agents in the physical world. [...]

Name

Family name: 馬(Mǎ) means horse.

Given name: 平(Píng) 川(chuān) means flat and level ground without geographical barrier.

Quoted from a Chinese idiom 一(yī) 馬(mǎ) 平(píng) 川(chuān) , my full name means the flat ground that one can ride straight across and thus implies enjoying a smooth life.

眾(zhòng) 峰(fēng) 來(lái) 自(zì) 天(tiān) 目(mù) 山(shān) ，勢(shì) 若(ruò) 駿(jùn) 馬(mǎ) 奔(bèn) 平(píng) 川(chuān) 。

宋(Sòng) · 蘇(Sū) 軾(Shì) 《東(Dōng) 坡(pō) 全(quán) 集(jí) · 卷(Juàn) 三(sān) · 遊(Yóu) 徑(jìng) 山(shān) 》

There are lots of mountains from Tianmu Mountain, which have the might of fine horses galloping across flat ground.

Su Shi of the Song dynasty in Visit Jing Mountain

Research

(* indicates equal contribution, highlight indicates representative papers)

LLM and Simulation as Bilevel Optimizers: A New Paradigm to Advance Physical Scientific Discovery

Pingchuan Ma, Tsun-Hsuan Wang, Minghao Guo, Zhiqing Sun, Joshua B. Tenenbaum, Daniela Rus, Chuang Gan, and Wojciech Matusik

ICML 2024 / Homepage / Paper / Code / bibtex

@inproceedings{ma2024llm,
  title={LLM and Simulation as Bilevel Optimizers: A New Paradigm to Advance Physical Scientific Discovery},
  author={Ma, Pingchuan and Wang, Tsun-Hsuan and Guo, Minghao and Sun, Zhiqing and Tenenbaum, Joshua B and Rus, Daniela and Gan, Chuang and Matusik, Wojciech},
  booktitle={International Conference on Machine Learning},
  year={2024},
  organization={PMLR}
}

DiffuseBot: Breeding Soft Robots with Physics-Augmented Generative Diffusion Models

Tsun-Hsuan Wang, Juntian Zheng, Pingchuan Ma, Yilun Du, Byungchul Kim, Andrew Spielberg, Joshua B. Tenenbaum, Chuang Gan, and Daniela Rus

NeurIPS 2023 / Homepage / Paper / Video / bibtex

Oral Presentation

@article{wang2024diffusebot,
  title={{DiffuseBot}: Breeding Soft Robots With Physics-Augmented Generative Diffusion Models},
  author={Wang, Tsun-Hsuan Johnson and Zheng, Juntian and Ma, Pingchuan and Du, Yilun and Kim, Byungchul and Spielberg, Andrew and Tenenbaum, Josh and Gan, Chuang and Rus, Daniela},
  journal={Advances in Neural Information Processing Systems},
  volume={36},
  year={2024}
}

Learning Neural Constitutive Laws from Motion Observations for Generalizable PDE Dynamics

Pingchuan Ma, Peter Yichen Chen, Bolei Deng, Joshua B. Tenenbaum, Tao Du, Chuang Gan, and Wojciech Matusik

ICML 2023 / Homepage / Paper / Code / bibtex

@inproceedings{ma2023learning,
  title={Learning Neural Constitutive Laws from Motion Observations for Generalizable PDE Dynamics},
  author={Ma, Pingchuan and Chen, Peter Yichen and Deng, Bolei and Tenenbaum, Joshua B and Du, Tao and Gan, Chuang and Matusik, Wojciech},
  booktitle={International Conference on Machine Learning},
  pages={23279--23300},
  year={2023},
  organization={PMLR}
}

SoftZoo: A Soft Robot Co-design Benchmark For Locomotion In Diverse Environments

Tsun-Hsuan Wang, Pingchuan Ma, Andrew Spielberg, Zhou Xian, Hao Zhang, Joshua B. Tenenbaum, Daniela Rus, and Chuang Gan

ICLR 2023 / Homepage / Paper / Code / bibtex

@inproceedings{wang2022softzoo,
  title={{SoftZoo}: A Soft Robot Co-design Benchmark for Locomotion in Diverse Environments},
  author={Wang, Tsun-Hsuan and Ma, Pingchuan and Spielberg, Andrew Everett and Xian, Zhou and Zhang, Hao and Tenenbaum, Joshua B and Rus, Daniela and Gan, Chuang},
  booktitle={The Eleventh International Conference on Learning Representations},
  year={2022}
}

RISP: Rendering-Invariant State Predictor with Differentiable Simulation and Rendering for Cross-Domain Parameter Estimation

Pingchuan Ma^*, Tao Du^*, Joshua B. Tenenbaum, Wojciech Matusik, and Chuang Gan

ICLR 2022 / Homepage / Paper / Demo / MIT News / bibtex

Oral Presentation

@inproceedings{ma2021risp,
  title={{RISP}: Rendering-Invariant State Predictor with Differentiable Simulation and Rendering for Cross-Domain Parameter Estimation},
  author={Ma, Pingchuan and Du, Tao and Tenenbaum, Joshua B and Matusik, Wojciech and Gan, Chuang},
  booktitle={International Conference on Learning Representations},
  year={2021}
}

DiffPD: Differentiable Projective Dynamics

Tao Du, Kui Wu, Pingchuan Ma, Sebastien Wah, Andrew Spielberg, and Wojciech Matusik

SIGGRAPH 2022 / Homepage / Paper / Code / Video / Slides / bibtex

@article{du2021diffpd,
  title={{DiffPD}: Differentiable Projective Dynamics},
  author={Du, Tao and Wu, Kui and Ma, Pingchuan and Wah, Sebastien and Spielberg, Andrew and Rus, Daniela and Matusik, Wojciech},
  journal={ACM Transactions on Graphics (TOG)},
  volume={41},
  number={2},
  pages={1--21},
  year={2021},
  publisher={ACM New York, NY}
}

DiffAqua: A Differentiable Computational Design Pipeline for Soft Underwater Swimmers with Shape Interpolation

Pingchuan Ma, Tao Du, John Z. Zhang, Kui Wu, Andrew Spielberg, Robert K. Katzschmann, and Wojciech Matusik

SIGGRAPH 2021 / Homepage / Paper / Code / Video / bibtex

@article{ma2021diffaqua,
  title={{DiffAqua}: A Differentiable Computational Design Pipeline for Soft Underwater Swimmers with Shape Interpolation},
  author={Ma, Pingchuan and Du, Tao and Zhang, John Z and Wu, Kui and Spielberg, Andrew and Katzschmann, Robert K and Matusik, Wojciech},
  journal={ACM Transactions on Graphics (TOG)},
  volume={40},
  number={4},
  pages={1--14},
  year={2021},
  publisher={ACM New York, NY, USA}
}

Efficient Continuous Pareto Exploration in Multi-Task Learning

Pingchuan Ma^*, Tao Du^*, and Wojciech Matusik

ICML 2020 / Homepage / Paper / Code / Talk / Slides / bibtex

@inproceedings{ma2020efficient,
  title={Efficient Continuous Pareto Exploration in Multi-Task Learning},
  author={Ma, Pingchuan and Du, Tao and Matusik, Wojciech},
  booktitle={International Conference on Machine Learning},
  pages={6522--6531},
  year={2020},
  organization={PMLR}
}

Prediction-Guided Multi-Objective Reinforcement Learning for Continuous Robot Control

Jie Xu, Yunsheng Tian, Pingchuan Ma, Daniela Rus, Shinjiro Sueda, and Wojciech Matusik

ICML 2020 / Homepage / Paper / Code / Video / bibtex

@inproceedings{xu2020prediction,
  title={Prediction-guided multi-objective reinforcement learning for continuous robot control},
  author={Xu, Jie and Tian, Yunsheng and Ma, Pingchuan and Rus, Daniela and Sueda, Shinjiro and Matusik, Wojciech},
  booktitle={International conference on machine learning},
  pages={10607--10616},
  year={2020},
  organization={PMLR}
}

Learning Efficient Video Representation with Video Shuffle Networks

Pingchuan Ma^*, Yao Zhou^*, Yu Lu, and Wei Zhang

arXiv 2019 / Paper / bibtex

@article{ma2019learning,
  title={Learning Efficient Video Representation with Video Shuffle Networks},
  author={Ma, Pingchuan and Zhou, Yao and Lu, Yu and Zhang, Wei},
  journal={arXiv preprint arXiv:1911.11319},
  year={2019}
}

Fluid Directed Rigid Body Control using Deep Reinforcement Learning

Pingchuan Ma^*, Yunsheng Tian^*, Zherong Pan, Bo Ren, and Dinesh Manocha

SIGGRAPH 2018 / Homepage / Paper / Video / bibtex

@article{ma2018fluid,
  title={Fluid Directed Rigid Body Control using Deep Reinforcement Learning},
  author={Ma, Pingchuan and Tian, Yunsheng and Pan, Zherong and Ren, Bo and Manocha, Dinesh},
  journal={ACM Transactions on Graphics (TOG)},
  volume={37},
  number={4},
  pages={1--11},
  year={2018},
  publisher={ACM New York, NY, USA}
}

Education

Massachusetts Institute of Technology

Ph.D. Candidate in Computer Science

Cambridge, MA

Sep. 2019 - Present

Massachusetts Institute of Technology

S.M. in Computer Science

Cambridge, MA

Sep. 2019 - Jan. 2023

Nankai University

B.Eng. in Software Engineering

Tianjin, China