Background

Bin Nie is a fourth-year Ph.D. candidate in the Department of Computer Science at the College of William and Mary, under the supervision of Prof. Evgenia Smirni. Before joining W&M, she received her bachelor degree in Software Engineering from Xiamen University in 2012, and master degree in Computer Science from Fordham University in 2014.

Skills

  • Programming Languages:  Python, Java, C/C++, SQL, Matlab, C#, Shell
  • Open Source Tools:  PyTorch, Scikit-learn, LibSVM/LibLinear, GPGPU-Sim

Professional Experience

  • 2017/5 - 2017/8 Research Intern, NEC Labs America Princeton, NJ
  • 2016/5 - 2016/8 Research Intern, Hewlett Packard Enterprise Palo Alto, CA
  • 2015/5 - 2015/8 Research Intern, Oak Ridge National Laboratory Oak Ridge, TN
  • 2014/8 - Present Research Assistant, College of William and Mary Williamsburg, VA
  • 2012/8 - 2014/5 Research Assistant, Fordham University New York, NY

Publications

Machine Learning Models for GPU Error Prediction in a Large Scale HPC System [PDF]

Bin Nie, Ji Xue, Saurabh Gupta, Tirthak Patel, Christian Engelmann, Evgenia Smirni, and Devesh Tiwari

In the 48th International Conference on Dependable Systems and Networks (DSN), Luxembourg City, Luxembourg, 2018

(Acceptance rate ≈ 26.7%, to appear)

Fill-in the Gaps: Spatial-Temporal Models for Missing Data [PDF]

Ji Xue, Bin Nie, and Evgenia Smirni

In the 13th International Conference on Network and Service Management (CNSM), Tokyo, Japan, November, 2017

(Acceptance rate ≈ 17.6%)

Characterizing Temperature, Power, and Soft-Error Behaviors in Data Center Systems: Insights, Challenges, and Opportunities [PDF]

Bin Nie, Ji Xue, Saurabh Gupta, Christian Engelmann, Evgenia Smirni, and Devesh Tiwari

In the 25th IEEE International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS), Banff, AB, Canada, September, 2017

(Acceptance rate ≈ 30.9%)

A Large-Scale Study of Soft-Errors on GPUs in the Field [PDF]

Bin Nie, Devesh Tiwari, Saurabh Gupta, Evgenia Smirni, and James H. Rogers

In the 22nd IEEE International Symposium on High Performance Computer Architecture (HPCA), Barcelona, Spain, March, 2016

(Acceptance rate ≈ 22%)

Leveraging Online Social Friendship to Improve Data Swarming Performance

Honggang Zhang, Benyuan Liu, Bin Nie, Zhiyong Xu, Xiayin Weng, and Chao Yu

Computer Networks, 2014

Social Interaction Based Video Recommendation: Recommending YouTube Videos to Facebook Users [PDF]

Bin Nie, Honggang Zhang, and Yong Liu

In 2014 Proceedings IEEE INFOCOM Workshops, Toronto, ON, Canada, April, 2014.

Research Projects

College of William and Mary August 2015 - Present

Research Assistant
Advisor: Prof. Evgenia Smirni
  • Analyzed the impact of temperature and power on GPU single-bit errors (SBEs) in the Titan supercomputer
  • Developed an ML-based error prediction model, resulting in 44% improvement in F1 Score for SBE prediction
  • Proposed a similarity-based sample reduction model to address imbalanced data set problem

System Research at NEC Labs America May 2017 - August 2017

Research Intern
Mentor: Dr. Jianwu Xu and Dr. Hui Zhang
  • Built a LSTM-based failure prediction model, named LoPA (Log Predictive Analytics), for large-scale heterogeneous system logs
  • Interpreted LoPA predictions with a local surrogate model (i.e., LIME)

Analytics Lab at Hewlett Packard Enterprise (HPE) May 2016 - August 2016

Research Intern
Mentor: Dr. Mehran Kafai and Dr. Kave Eshghi
  • Compared the effect of various feature engineering and ML models on click-through rate (CTR) prediction
  • Built a linear model by applying CROHash+LR on TF-IDF weighted categorical data

Oak Ridge National Laboratory (ORNL) May 2015 - August 2015

Research Intern
Mentor: Dr. Devesh Tiwari
  • Conducted a large-scale study of soft-errors (i.e., SBE, DBE, DPR) on GPU nodes in the Titan* supercomputer
  • Investigated performance variance on soft-error-affected and error-free GPU nodes
  • Studied how user/application behaviors changed over time and their indications on errors
  • * Titan supercomputer, hosted in ORNL, is the world's second fastest supercomputer (as of Nov 2015 Top500 Supercomputer rankings).

Fordham University August 2012 - May 2014

Research Assistant
Advisor: Prof. Honggang Zhang
  • Conducted a measurement study of how social interactions on Facebook relate to video popularity on YouTube
  • Designed a social interaction based video recommendation algorithm, yielding to 20%~25% improvement in accuracy
  • Built a crawling platform (for data retrieval from Facebook & Twitter) and social network analysis modules

TA Experience

  • CSCI420 Human Computer Interface Design --- Spring 2016
  • CSCI426 Simulation --- Fall 2015
  • CSCI141 Computational Problem Solving in Python --- Fall 2014, Spring 2015

Awards

  • Student Travel Grant for CNSM 2017, awarded by IFIP --- 2017
  • Student Travel Grant for HPCA 2016, awarded by IEEE --- 2016
  • OGSR/Graduate Student Association Conference Funds, awarded by College of William and Mary --- 2016
  • Conference Travel Fund Award, awarded by College of William and Mary --- 2016
  • GHC (Grace Hopper Celebration of Women in Computing) Scholarship --- 2015
  • Graduate Studies Advisory Board Fellowship, awarded by W&M Graduate Studies Advisory Board --- 2014, 2015
  • National Scholarship, awarded to top 2% undergraduate students by Ministry of Education of China --- 2010

Last updated in November, 2017