Proposal documents


Short Bio

I am near the end of my fourth year. CMU really introduced me to machine learning and I am in deep love with it. I work on active learning --- more specifically: active search of the positive class with complex actions and rewards. To apply active search to more problems, I assume different probabilistic graphical models between collectable variables and reward patterns. The methods that I consider generally belong to the families of Bayesian optimization and multi-armed bandits. For the theory part, I am interested in regret analysis, spectral graph theories, as well as diversity via submodular set functions. I also want to explore online learning and distributed machine learning, which both include an automated decision process about what to prioritize next. After all, they also connect to big data more naturally.





Publications


  • Yifei Ma, Dougal J. Sutherland, Roman Garnett, Jeff Schneider. Active Pointillistic Pattern Search. AISTATS 2015. Two Shared Lead Authors.
  • Yifei Ma, Dougal J. Sutherland, Roman Garnett, Jeff Schneider. Active Pointillistic Pattern Search. NIPS 2014 Workshop on Bayesian Optimization.
  • Yifei Ma, Roman Garnett, Jeff Schneider. Active Area Search via Bayesian Quadrature. AISTATS 2014.
  • Yifei Ma, Roman Garnett, Jeff Schneider. Sigma-Optimality for Active Learning on Gaussian Random Fields. NIPS 2013.
  • Guangyu Xia, Tongbo Huang, Yifei Ma, Roger B. Dannenberg, Christos Faloutsos. MidiFind: Similarity Search and Popularity Mining in Large MIDI Databases. CMMR 2013: 259-276.
  • Yifei Ma, Roman Garnett, Jeff Schneider. Submodularity in Batch Active Learning and Survey Problems on Gaussian Random Fields. NIPS 2012 Workshop on DISCML.
  • Yifei Ma, Li Li, Xiaolin Huang, Shuning Wang, Robust Support Vector Machine Using Least Median Loss Penalty, Proceedings of the 18th IFAC World Congress, Volume 18, Part 1, 2011. [Submitted pdf][Details]

Experience


Citadel LLC. Quantitative Researcher Intern. 5/2014-8/2014 at Chicago.

  • Feature design and analysis for market value prediction.
  • Bayesian methods for low-frequency predictions.

Education


Ph.D. student in Machine Learning Dept., School of Computer Science, Carnegie Mellon University, 8/2011-now.

B.S. in Automation, Dual B.S. in Mathematics, Tsinghua University, 8/2007-7/2011.

  • 1st GPA in Automation by the time of graduate school application.
  • Totally 11 students dual majored in Mathematics in a class of 3000+ students.

Exchange Study (Credits Transferred), Georgia Inst. of Technology, 8/2009-12/2009.

  • A rare opportunity given to only 6 students university-wide.

Honors and Awards

2007-2010, 1st Level Academic Scholarship every year in Tsinghua University;
10/2009, 31/700 in IEEEXtreme 24-Hour Programming Competition (team leader);
10/2006, 1st prize in National High School Physics Competition;
03/2006, Qualified for the American Invitational Mathematics Examination (no further award for foreigners);
10/2004, 1st prize in National High School Math Competition (over 2 years younger than most others).

Volunteer Experience

  • Joined student volunteer group for over 10 years;
  • Served as a regular campus guide for visitors in Tsinghua University;
  • Volunteered for the 2008 Beijing Olympic Games (in medal ceremonies as a winning national flag raiser).

Student Activities and Hobbies

  • Editor for Tsinghua Student Psychology Club Journal;
  • Member of Campus Tour Guide Association;
  • Member of Tsinghua Honor Guard (to display and escort the national flag on ceremonial occasions);
  • Volunteer teaching assistant for 30 students on Fundamentals of Analog Electronics and Data Structure;
  • Classical Chinese Philosophy like the Analects (Confucius) and Tao Te Qing (Lao-tzu);
  • Piano, swimming, and running.
Photo of me.

Animation powered by jQuery and its plugins. Design idea partially stolen from Ye Tao.