Publications

Conference Papers (Dr. Lin’s students; * equal contribution)

  1. Kaidi Xu, Gaoyuan Zhang, Sijia Liu, Quanfu Fan, Meng Sun, Hongge Chen, Pin-Yu Chen, Yanzhi Wang, and Xue Lin. Evading Real-Time Person Detectors by Adversarial T-shirt. In arXiv, Oct 2019.
  2. [NDSS’20][Best Technical Poster Award] Takami Sato, Junjie Shen, Ningfei Wang, Yunhan Jia, Xue Lin, and Qi Chen. Security of deep learning based lane keeping assistance systems under physical-world adversarial attack. Presented at the NDSS Symposium, Feb 23 – 26, 2020 in San Diego, California.
  3. [DAC’20] Mengshu Sun, Pu Zhao, Mehmet Gungor, Miriam Leeser, Massoud Pedram, and Xue Lin. 3D CNN acceleration on FPGA using hardware-aware pruning. In Proceedings of the 57th Annual Design Automation Conference 2020, page XXX. ACM, 2020. Acceptance rate: 23% 
  4. [DAC’20] Peiyan Dong, Siyue Wang, Wei Niu, Chengming Zhang, Sheng Lin, Zhengang Li, Yifan Gong, Bin Ren, Xue Lin, and Dingwen Tao. RTMobile: beyond real-time mobile acceleration of RNNs for speech recognition. In Proceedings of the 57th Annual Design Automation Conference 2020, page XXX. ACM, 2020. Acceptance rate: 23% 
  5. [ICASSP’20Kaidi Xu, Sijia Liu, Pin-Yu Chen, Mengshu Sun, Caiwen Ding, Bhavya Kailkhura, and Xue Lin. Towards an efficient and general framework of robust training for graph neural networks. In Proceedings of the ICASSP, 2020.
  6. [ICASSP’20] Xiao Wang, Siyue Wang, Pin-Yu Chen, Xue Lin, and Peter Chin. ADVMS: a multi-source multi-cost defense against adversarial attacks. In Proceedings of the ICASSP, 2020.
  7. [ASPLOS’20] Wei Niu, Xiaolong Ma, Sheng Lin, Shihao Wang, Xuehai Qian, Xue Lin, Yanzhi Wang, and Bin Ren. PatDNN: achieving real-time DNN execution on mobile devices with pattern-based weight pruning. In Proceedings of the Twenty-Fifth International Conference on Architectural Support for Programming Languages and Operating Systems. ACM, 2020. Acceptance rate: 21.1%
  8. [ICLR’20Pu Zhao, Pin-Yu Chen, Payel Das, Karthikeyan Natesan Ramamurthy, and Xue Lin. Bridging mode connectivity in loss landscapes and adversarial robustness. In International Conference on Learning Representations, 2020. Acceptance rate: 26.5% (687/2594)
  9. [AAAI’20Pu Zhao, Pin-Yu Chen, Siyue Wang, and Xue Lin. Towards query-efficient black-box adversary with zeroth-order natural gradient descent. In Proceedings of the AAAI Conference on Artificial Intelligence, 2020. Acceptance rate: 20.6% (1591/7737)
  10. [AAAI’20] Lily Weng*, Pu Zhao*, Sijia Liu, Pin-Yu Chen, Xue Lin, and Luca Daniel. Towards certificated model robustness against weight perturbations. In Proceedings of the AAAI Conference on Artificial Intelligence, 2020. Acceptance rate: 20.6% (1591/7737)
  11. [AAAI’20] Xiaolong Ma, Fu-ming Guo, Wei Niu, Xue Lin, Jian Tang, Bin Ren, and Yanzhi Wang. PCONV: the missing but desirable sparsity in DNN weight pruning for real-time execution on mobile device. In Proceedings of the AAAI Conference on Artificial Intelligence, 2020. Acceptance rate: 20.6% (1591/7737)
  12. [HOST’20 Tutorial] Xue Lin, Yunsi Fei, and Thomas Wahl. Protecting confidentiality and integrity of deep neural networks. In IEEE International Symposium on Hardware Oriented Security and Trust (HOST), 2020.
  13. [NeurIPS’19] Xiangyi Chen*, Sijia Liu*, Kaidi Xu*, Xingguo Li, Xue Lin, Mingyi Hong, and David Cox. Zo-adamm: Zeroth-order adaptive momentum method for black-box optimization. In Advances in Neural Information Processing Systems, 2019. Acceptance rate: 21.1% (1428/6743)
  14. [ICCV’19Pu Zhao, Sijia Liu, Pin-Yu Chen, Nghia Hoang, Kaidi Xu, Bhavya Kailkhura, and Xue Lin. On the design of black-box adversarial examples by leveraging gradient-free optimization and operator splitting method. In Proceedings of the IEEE International Conference on Computer Vision, pages 121–130, 2019. Acceptance rate: 25% (1077/4303)
  15. [ICCV’19] Shaokai Ye*, Kaidi Xu*, Sijia Liu, Hao Cheng, Jan-Henrik Lambrechts, Huan Zhang, Aojun Zhou, Kaisheng Ma, Yanzhi Wang, and Xue Lin. Adversarial robustness vs model compression, or both? In Proceedings of the IEEE International Conference on Computer Vision, pages 111–120, 2019. Acceptance rate: 25% (1077/4303)
  16. [IJCAI’19Kaidi Xu, Hongge Chen, Sijia Liu, Pin-Yu Chen, Tsui-Wei Weng, Mingyi Hong, and Xue Lin. Topology attack and defense for graph neural networks: An optimization perspective. In Proceedings of the 28th International Joint Conference on Artificial Intelligence. AAAI Press, 2019. Acceptance rate: 17.8% (850/4752)
  17. [IJCAI’19] Xiao Wang*, Siyue Wang*, Pin-Yu Chen, Yanzhi Wang, Brian Kulis, Xue Lin, and Peter Chin. Protecting neural networks with hierarchical random switching: towards better robustness-accuracy trade-off for stochastic defenses. In Proceedings of the 28th International Joint Conference on Artificial Intelligence, pages 6013–6019. AAAI Press, 2019. Acceptance rate: 17.8% (850/4752)
  18. [CVPR’19] Zihao Liu, Qi Liu, Tao Liu, Nuo Xu, Xue Lin, Yanzhi Wang, and Wujie Wen. Feature distillation: Dnn-oriented jpeg compression against adversarial examples. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 860–868, 2019. Acceptance rate: 25.2% (1299/5165)
  19. Hao Cheng, Kaidi Xu, Sijia Liu, Pin-Yu Chen, Pu Zhao, and Xue Lin. Defending against Backdoor Attack on Deep Neural Networks. KDD Workshop on Adversarial Learning Methods for Machine Learning and Data Mining (AdvML), 2019.
  20. Xiao Wang, Siyue Wang, Pin-Yu Chen, Xue Lin, and Peter Chin. Block Switching: A Stochastic Approach for Deep Learning Security. KDD Workshop on Adversarial Learning Methods for Machine Learning and Data Mining (AdvML), 2019.
  21. [GLSVLSI’19Mengshu Sun, Pu Zhao, Yanzhi Wang, Naehyuck Chang, and Xue Lin. Hsim-dnn: Hardware simulator for computation-, storage-and power-efficient deep neural networks. In Proceedings of the 2019 on Great Lakes Symposium on VLSI, pages 81–86. ACM, 2019. Acceptance rate: 29%
  22. [DAC’19Pu Zhao, Siyue Wang, Cheng Gongye, Yanzhi Wang, Yunsi Fei, and Xue Lin. Fault sneaking attack: a stealthy framework for misleading deep neural networks. In Proceedings of the 56th Annual Design Automation Conference 2019, page 165. ACM, 2019. Acceptance rate: 24.8% (202/815)
  23. [ICLR’19Kaidi Xu, Sijia Liu, Pu Zhao, Pin-Yu Chen, Huan Zhang, Quanfu Fan, Deniz Erdogmus, Yanzhi Wang, and Xue Lin. Structured adversarial attack: Towards general implementation and better interpretability. In International Conference on Learning Representations, 2019. Acceptance rate: 31% (500/1591)
  24. [ASPLOS’19] Ao Ren, Tianyun Zhang, Shaokai Ye, Jiayu Li, Wenyao Xu, Xuehai Qian, Xue Lin, and Yanzhi Wang. Admm-nn: An algorithm-hardware co-design framework of dnns using alternating direction methods of multipliers. In Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems, pages 925–938. ACM, 2019. Acceptance rate: 21.1% (74/350)
  25. [HPCA’19] Zhe Li, Caiwen Ding, Siyue Wang, Wujie Wen, Youwei Zhuo, Chang Liu, Qinru Qiu, Wenyao Xu, Xue Lin, Xuehai Qian, and Yanzhi Wang. E-rnn: Design optimization for efficient recurrent neural networks in fpgas. In 2019 IEEE International Symposium on High Performance Computer Architecture, pages 69–80. IEEE, 2019. Acceptance rate: 19.7% (46/233)
  26. [ASP-DAC’19Pu Zhao, Kaidi Xu, Sijia Liu, Yanzhi Wang, and Xue Lin. Admm attack: an enhanced adversarial attack for deep neural networks with undetectable distortions. In Proceedings of the 24th Asia and South Pacific Design Automation Conference, pages 499–505. ACM, 2019.
  27. [AAAI’19] Yanzhi Wang, Zheng Zhan, Liang Zhao, Jian Tang, Siyue Wang, Jiayu Li, Bo Yuan, Wujie Wen, and Xue Lin. Universal approximation property and equivalence of stochastic computing-based neural networks and binary neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 5369–5376, 2019. Acceptance rate: 16.2% (1150/7095)
  28. Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, and Xue Lin. Defending DNN adversarial attacks with pruning and logits augmentation. In Proc. of IEEE GlobalSIP 2018, Nov. 2018.
  29. Pu Zhao, Kaidi Xu, Tianyun Zhang, Markan Fardad, Yanzhi Wang, and Xue Lin. Reinforced adversarial attacks on deep neural networks using ADMM. In Proc. of IEEE GlobalSIP 2018, Nov. 2018.
  30. [ACM MM’18Pu Zhao, Sijia Liu, Yanzhi Wang, and Xue Lin. An admm-based universal framework for adversarial attacks on deep neural networks. In 2018 ACM Multimedia Conference on Multimedia Conference, pages 1065–1073. ACM, 2018. Acceptance rate: 27.5% (144/757)
  31. [ICCAD’18][Best Paper NominationSiyue Wang, Xiao Wang, Pu Zhao, Wujie Wen, David Kaeli, Peter Chin, and Xue Lin. Defensive dropout for hardening deep neural networks under adversarial attacks. In Proceedings of the International Conference on Computer-Aided Design, page 71. ACM, 2018. Acceptance rate: 25% (98/396)
  32. [AAAI’18] Yanzhi Wang, Caiwen Ding, Zhe Li, Geng Yuan, Siyu Liao, Xiaolong Ma, Bo Yuan, Xuehai Qian, Jian Tang, Qinru Qiu, and Xue Lin. Towards ultra-high performance and energy efficiency of deep learning systems: an algorithm-hardware co-optimization framework. In Thirty-Second AAAI Conference on Artificial Intelligence, 2018.
  33. [ASP-DAC’18Pu Zhao, Yanzhi Wang, Naehyuck Chang, Qi Zhu, and Xue Lin. A deep reinforcement learning framework for optimizing fuel economy of hybrid electric vehicles. In 2018 23rd Asia and South Pacific Design Automation Conference, pages 196–202. IEEE, 2018.
  34. [ICCAD’17] Siyu Liao, Zhe Li, Xue Lin, Qinru Qiu, Yanzhi Wang, and Bo Yuan. Energy-efficient, high-performance, highly-compressed deep neural network design using block-circulant matrices. In 2017 IEEE/ACM International Conference on Computer-Aided Design, pages 458–465. IEEE, 2017.
  35. [MICRO’17] Caiwen Ding, Siyu Liao, Yanzhi Wang, Zhe Li, Ning Liu, Youwei Zhuo, Chao Wang, Xuehai Qian, Yu Bai, Geng Yuan, Jian Tang, Qinru Qiu, Xue Lin, and Bo Yuan. C ir cnn: accelerating and compressing deep neural networks using block-circulant weight matrices. In Proceedings of the 50th Annual IEEE/ACM International Symposium on Microarchitecture, pages 395–408. ACM, 2017.
  36. [ASP-DAC’17] Caiwen Ding, Ji Li, Weiwei Zheng, Naehyuck Chang, Xue Lin, and Yanzhi Wang. Algorithm accelerations for luminescent solar concentrator-enhanced reconfigurable onboard photovoltaic system. In 2017 22nd Asia and South Pacific Design Automation Conference, pages 318–323. IEEE, 2017.
  37. [ICCD’16] Xue Lin, Yuankun Xue, Paul Bogdan, Yanzhi Wang, Siddharth Garg, and Massoud Pedram. Power-aware virtual machine mapping in the data-center-on-a-chip paradigm. In 2016 IEEE 34th International Conference on Computer Design, pages 241–248. IEEE, 2016.
  38. [ICCD’16] Caiwen Ding, Hongjia Li, Weiwei Zheng, Yanzhi Wang, Naehyuck Chang, and Xue Lin. Luminescent solar concentrator-based photovoltaic reconfiguration for hybrid and plug-in electric vehicles. In 2016 IEEE 34th International Conference on Computer Design, pages 281–288. IEEE, 2016.
  39. [CLOUD’16] Xue Lin, Massoud Pedram, Jian Tang, and Yanzhi Wang. A profit optimization framework of energy storage devices in data centers: Hierarchical structure and hybrid types. In 2016 IEEE 9th International Conference on Cloud Computing, pages 640–647. IEEE, 2016.
  40. [ICCAD’15] Xue Lin, Paul Bogdan, Naehyuck Chang, and Massoud Pedram. Machine learning-based energy management in a hybrid electric vehicle to minimize total operating cost. In 2015 IEEE/ACM International Conference on Computer-Aided Design, pages 627–634. IEEE, 2015.
  41. [DAC’15] Yanzhi Wang, Xue Lin, Massoud Pedram, and Naehyuck Chang. Joint automatic control of the powertrain and auxiliary systems to enhance the electromobility in hybrid electric vehicles. In 2015 52nd ACM/EDAC/IEEE Design Automation Conference, pages 1–6. IEEE, 2015.
  42. [DATE’15] Xue Lin, Yanzhi Wang, Massoud Pedram, Jaemin Kim, and Naehyuck Chang. Event-driven and sensorless photovoltaic system reconfiguration for electric vehicles. In Proceedings of the 2015 Design, Automation & Test in Europe Conference & Exhibition, pages 19–24. EDA Consortium, 2015.
  43. [ICCAD’14] Xue Lin, Yanzhi Wang, Paul Bogdan, Naehyuck Chang, and Massoud Pedram. Reinforcement learning based power management for hybrid electric vehicles. In Proceedings of the 2014 IEEE/ACM International Conference on Computer-Aided Design, pages 32–38. IEEE Press, 2014.
  44. [ICCD’14] Xue Lin, Yanzhi Wang, Naehyuck Chang, and Massoud Pedram. Power supply and consumption co-optimization of portable embedded systems with hybrid power supply. In 2014 IEEE 32nd International Conference on Computer Design, pages 477–482. IEEE, 2014.
  45. [ISVLSI’14][Best Paper Award] Alireza Shafaei, Yanzhi Wang, Xue Lin, and Massoud Pedram. Fincacti: Architectural analysis and modeling of caches with deeply-scaled finfet devices. In 2014 IEEE Computer Society Annual Symposium on VLSI, pages 290–295. IEEE, 2014.
  46. [CLOUD’14][Top Paper Award] Xue Lin, Yanzhi Wang, Qing Xie, and Massoud Pedram. Energy and performance-aware task scheduling in a mobile cloud computing environment. In 2014 IEEE 7th International Conference on Cloud Computing, pages 192–199. IEEE, 2014.
  47. [ICCAD’13] Xue Lin, Yanzhi Wang, and Massoud Pedram. Joint sizing and adaptive independent gate control for finfet circuits operating in multiple voltage regimes using the logical effort method. In 2013 IEEE/ACM International Conference on Computer-Aided Design, pages 444–449. IEEE, 2013.
  48. [ISLPED’13] Xue Lin, Yanzhi Wang, Siyu Yue, Naehyuck Chang, and Massoud Pedram. A framework of concurrent task scheduling and dynamic voltage and frequency scaling in real-time embedded systems with energy harvesting. In International Symposium on Low Power Electronics and Design, pages 70–75. IEEE, 2013.
  49. [ICCAD’12] Xue Lin, Yanzhi Wang, Di Zhu, Naehyuck Chang, and Massoud Pedram. Online fault detection and tolerance for photovoltaic energy harvesting systems. In Proceedings of the International Conference on Computer-Aided Design, pages 1–6. ACM, 2012.
  50. [DAC’12] Xue Lin, Yanzhi Wang, Siyu Yue, Donghwa Shin, Naehyuck Chang, and Massoud Pedram. Near-optimal, dynamic module reconfiguration in a photovoltaic system to combat partial shading effects. In DAC Design Automation Conference 2012, pages 516–521. IEEE, 2012.

Journal Papers

  1. [Submitted] Yanzhi Wang, Shaokai Ye, Zhezhi He, Xiaolong Ma, Linfeng Zhang, Sheng Lin, Geng Yuan, Sia Huat Tan, Zhengang Li, Deliang Fan, Xuehai Qian, Xue Lin, and Kaisheng Ma. Non-structured DNN weight pruning considered harmful. arXiv:1907.02124, 2019.
  2. [PloS one’18Mengshu Sun, Yuankun Xue, Paul Bogdan, Jian Tang, Yanzhi Wang, and Xue Lin. Hierarchical and hybrid energy storage devices in data centers: Architecture, control and provisioning. PloS one, 13(1):e0191450, 2018.
  3. [T VLSI’18] Jaemin Kim, Donkyu Baek, Caiwen Ding, Sheng Lin, Donghwa Shin, Xue Lin, Yanzhi Wang, Young Hoo Cho, Sang Hyun Park, and Naehyuck Chang. Dynamic reconfiguration of thermoelectric generators for vehicle radiators energy harvesting under location-dependent temperature variations. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 26(7):1241–1253, 2018.
  4. [Design&Test’18] Caiwen Ding, Hongjia Li, Weiwei Zheng, Yanzhi Wang, and Xue Lin. Reconfigurable photovoltaic systems for electric vehicles. IEEE Design & Test, 35(6):37–43, 2018.
  5. [IET CPS’17Pu ZhaoXue Lin, Yanzhi Wang, Shuang Chen, and Massoud Pedram. Hierarchical resource allocation and consolidation framework in a multi-core server cluster using a markov decision process model. IET Cyber-Physical Systems: Theory & Applications, 2(3):118–126, 2017.
  6. [T SC’17] Siyu Liao, Yi Xie, Xue Lin, Yanzhi Wang, Min Zhang, and Bo Yuan. Reduced-complexity deep neural networks design using multi-level compression. IEEE Transactions on Sustainable Computing, 2017.
  7. [T CAD’16] Xue Lin, Yanzhi Wang, Naehyuck Chang, and Massoud Pedram. Concurrent task scheduling and dynamic voltage and frequency scaling in a real-time embedded system with energy harvesting. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 35(11):1890–1902, 2016.
  8. [T SE’15] Yanzhi Wang, Xue Lin, and Massoud Pedram. A near-optimal model-based control algorithm for households equipped with residential photovoltaic power generation and energy storage systems. IEEE Transactions on Sustainable Energy, 7(1):77–86, 2015.
  9. [T CAS-II’15] Qing Xie, Xue Lin, Yanzhi Wang, Shuang Chen, Mohammad Javad Dousti, and Massoud Pedram. Performance comparisons between 7-nm finfet and conventional bulk cmos standard cell libraries. IEEE Transactions on Circuits and Systems II: Express Briefs, 62(8):761–765, 2015.
  10. [T SC’14] Xue Lin, Yanzhi Wang, Qing Xie, and Massoud Pedram. Task scheduling with dynamic voltage and frequency scaling for energy minimization in the mobile cloud computing environment. IEEE Transactions on Services Computing, 8(2):175–186, 2014.
  11. [T EC’14] Yanzhi Wang, Xue Lin, and Massoud Pedram. A stackelberg game-based optimization framework of the smart grid with distributed pv power generations and data centers. IEEE Transactions on Energy Conversion, 29(4):978–987, 2014.
  12. [T VLSI’14] Yanzhi Wang, Xue Lin, Younghyun Kim, Qing Xie, Massoud Pedram, and Naehyuck Chang. Single-source, single-destination charge migration in hybrid electrical energy storage systems. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 22(12):2752–2765, 2014.
  13. [T CAD’14] Yanzhi Wang, Xue Lin, Younghyun Kim, Naehyuck Chang, and Massoud Pedram. Architecture and control algorithms for combating partial shading in photovoltaic systems. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 33(6):917–930, 2014.
  14. [T Smart Grid’14] Yanzhi Wang, Xue Lin, and Massoud Pedram. Adaptive control for energy storage systems in households with photovoltaic modules. IEEE Transactions on Smart Grid, 5(2):992–1001, 2014.
  15. [Design&Test’13] Xue Lin, Yanzhi Wang, Massoud Pedram, Jaemin Kim, and Naehyuck Chang. Designing fault-tolerant photovoltaic systems. IEEE Design & Test, 31(3):76–84, 2013.
  16. [ACS Nano’12] Yuchi Che, Chuan Wang, Jia Liu, Bilu Liu, Xue Lin, Jason Parker, Cara Beasley, H-S Philip Wong, and Chongwu Zhou. Selective synthesis and device applications of semiconducting single-walled carbon nanotubes using isopropyl alcohol as feedstock. Acs Nano, 6(8):7454–7462, 2012.
  17. [Nano Research’10] Chuan Wang, Koungmin Ryu, Lewis Gomez De Arco, Alexander Badmaev, Jialu Zhang, Xue Lin, Yuchi Che, and Chongwu Zhou. Synthesis and device applications of high-density aligned carbon nanotubes using low-pressure chemical vapor deposition and stacked multiple transfer. Nano Research, 3(12):831–842, 2010.