I am a senior undergraduate student in Electrical Engineering Department at Princeton University. I am from Bangkok, Thailand. My name in Thai is "ชวิน สีตวาริน".
Currently, I am broadly interested in the fields of machine learning and computer security. I have worked on multiple research projects with Prof. Prateek Mittal, and my senior thesis advisor is Prof. Peter Ramadge. I also used to work on photonics with Prof. Alejandro Rodriguez.
Machine Learning and Security
- Physical attacks on traffic sign recognition systems
- C. Sitawarin, A. N. Bhagoji, A. Mosenia, P. Mittal, "Rogue Signs: Deceiving Traffic Sign Recognition with Malicious Ads and Logos," arXiv:1801.02780 (2018). (under review by IEEE S&P Deep Learning and Security Workshop) [paper]
- C. Sitawarin, A. N. Bhagoji, A. Mosenia, M. Chiang, P. Mittal, "DARTS: Deceiving Autonomous Cars with Toxic Signs," arXiv.:1802.06430 (2018). (under review by Ubicomp) [paper] [code]
- A. N. Bhagoji, D. Cullina, C. Sitawarin, P. Mittal, “Enhancing Robustness of Machine Learning System vis Data Transformations,” arXiv:1704.02654 (2017). [paper] [code]
- M. A. Martinez, C. Sitawarin, K. Finch, L. Meincke, A. Yablonski, A. Kornhauser, “Beyond Grand Theft Auto V for Training, Testing and Enhancing Deep Learning in Self Driving Cars,” arXiv:1712.01397 (2017). [paper] [webpage]
- C. Sitawarin, Z. Lin, W. Jin and A. W. Rodriguez, “Inverse designed photonic fibers and metasurfaces for nonlinear frequency conversion,” arXiv:1711.07810 (2017). (under review by Photonics Research) [paper]
- Z. Lin, C. Sitawarin, M. Lončar, A. W. Rodriguez, "Inverse-designed nonlinear nanophotonic structures: Enhanced frequency conversion at the nano scale," 2016 Conference on Lasers and Electro-Optics (CLEO), San Jose, CA, 2016, pp. 1-2. [paper]
GAN and Adversarial Examples
Senior thesis advised by Prof. Ramadge (2017 - Present)
We aim to explore and model the distribution of adversarial examples in the input space as well as construct novel attacks and defenses using variations of GAN
Physical Attacks on Traffic Sign Recognition Systems
We propose a range of real-world attacks against traffic sign recognition and detection systems of self-driving cars as well as possible defenses. We improve one existing attack based on adversarial examples and propose two novel attacks. Learn more
Practical Attack on Traffic Sign Recognition Systems
We construct practical adversarial examples for traffic sign classifiers and evaluate them under different lighting conditions and perspective transformations.
Defense Against PDF Malware Detector Evasion
We survey several evasion attacks on PDF malware detectors that rely on ML classifiers. Our experiment aims to find robust ML classifiers against such attacks and possible defenses including dimensionality reduction and adversarial training.
Summer Research Intern
Hong Kong Applied Science and Technology Research Institute (ASTRI) Hong Kong (Summer 2016)
I was a summer intern at ASTRI and a part of IC Digital Design team. My team focused on implementing a real-time super-resolution CNN on an FPGA. My work involved transferring the module written in C and Matlab and implementing it in RTL using Vivado High-Level Synthesis tool.
Lab TA and Grader
Electrical Engineering Department, Computer Science Department, Princeton University Princeton, NJ
- Contemporary Logic Design (ELE 206/COS 306) - Lab Teaching Assistant (Fall 2016)
- Information Security (COS/ELE 432) - Grader (Fall 2016)
- Algorithms and Data Structures (COS 226) - Grader (Spring 2016)
- General Computer Science (COS 126) - Grader (Fall 2015)
Awards & Honors
- The P. Michael Lion III Fund
- Shapiro Prize for Academic Excellence
- Tau Beta Pi - The Engineering Honor Society
- King’s Scholarship