Email:	akyrola ---- at ---- cs.cmu.edu
LinkedIn:	profile
Twitter:	kyrpov
I have graduated and joined Facebook!

I was a Ph.D. student (2009-2014) at the CMU Computer Science Department. My advisors are Carlos Guestin and Guy Blelloch.

My CV. Company I co-founded back in year 2000: Sulake Corporation. Moves is an automatic activity and location tracker for iOS and Android. I was co-founder of the company which was acquired by Facebook in April 2014.

Thesis

Thesis PDF. Thesis defense slides: pptx pdf

Talks

• Thesis defense slides: pptx pdf
• I gave keynote at the first Large-Scale Recommender System workshop at ACM RecSys'13: slides (powerpoint), slideshare

Publications (preprint)

• Aapo Kyrola, Carlos Guestrin: GraphChi-DB: Simple Design for a Scalable Graph System -- on Just a PC. Source code: graphchiDB-scala

Publications (peer-reviewed)

• Aapo Kyrola, Julian Shun, Guy Belloch: Beyond Synchronous: : New Techniques for External-Memory Graph Connectivity and Minimum Spanning Forest. (To be presented in SEA 2014)

• Aapo Kyrola: DrunkardMob: Billions of Random Walks on Just a PC (ACM RecSys 2013, Hong Kong)
• Download PDF

• Aapo Kyrola, G. Blelloch, C. Guestrin. GraphChi: Large-Scale Graph Computation on Just a PC. OSDI'12, Hollywood, CA.
• Download PDF
• OSDI talk slides
• GraphChi homepage
• MIT Technology Review article

• Y. Low, J. Gonzalez, A. Kyrola, D. Bickson, C. Guestrin, and J.M. Hellerstein. Distributed GraphLab: A Framework for Machine Learning in the Cloud. PVLDB 5(8):716-727, 2012.
• Download PDF
• GraphLab webpage

• R. Cheng, J. Hong, Aapo Kyrola, Y. Miao, X. Weng, M. Wu, F. Yang, L.Zhou, F. Zhao, E. Chen: Kineograph: Taking the Pulse of a Fast-Changing and Connected World. EUROSYS 2012.
• Download PDF

• Joseph Bradley*, Aapo Kyrola*, Danny Bickson, Carlos Guestrin. Parallel Coordinate Descent for L1-Regularized Loss Minimization. PDF
• Project page (incl. source code)

• Yucheng Low, Joseph Gonzalez, Aapo Kyrola, Danny Bickson, Carlos Guestrin. GraphLab: A New Framework for Parallel Machine Learning. Conference on Uncertainty in Artificial Intelligence (UAI), 2010. PDF
• GraphLab webpage

Projects and Software

• GraphChi webpage. Source code (GitHub)
• GraphChi-DB (GitHub)
• GraphLab webpage. GraphLab is a highly efficient parallel framework for Machine Learning in the multicore and distributed settings.
• Shotgun: parallel coordinate descent algorithm for L1-regularized linear and logistic regression.
• Parallel LDA Gibbs sampler (class project for Statistical Machine Learning): report code (Java).

• Callahan-Kosaraju algorithm for nearest neighbor search (class project for Algorithms in the Real World): report code (C++, CILK++).

• Simple Splay-tree applet: Splay Applet

Links: Linkedin   Moves app for automatic exercise tracking.