ANDREW DELONG Postdoctoral Fellow, Machine Learning & Inference www.psi.toronto.edu/∼ andrew andrew.delong@gmail.com Dept. of Electrical & Computer Engineering, University of Toronto, 10 King’s College Rd, Toronto, Ontario, Canada, M5S 3G4 EDUCATION 2011 PhD in Computer Science, University of Western Ontario. Thesis: Advances in Graph-Cut Optimization: Multi-Surface Models, Label Costs, and Hierarchical Costs (CIPPRS Dissertation Award) Advisor: Yuri Boykov 2006 MSc in Computer Science, University of Western Ontario. Thesis: A Scalable Max-Flow/Min-Cut Algorithm for Sparse Graphs Advisor: Yuri Boykov 2003 BMath in Honours Co-op Computer Science (87% grade average), University of Waterloo. AWARDS and DISTINCTIONS 2015–2016 Heffernan Postdoctoral Fellowship (5 recipients in Faculty of Engineering). 2012–2014 Postdoctoral Fellowship, National Sciences and Engineering Research Council (NSERC). — $80,000 over two years, awarded to 98 of 1254 applicants nationwide (8% success rate) 2011 Doctoral Dissertation Award, Canadian Image Processing and Pattern Recognition Society. 2010–2011 Ontario Graduate Scholarship in Science and Technology ($10,000). 2008 Faculty of Science Graduate Student Teaching Award, University of Western Ontario. 2007–2010 Postgraduate Scholarship, National Sciences and Engineering Research Council (NSERC). — $63,000 over three years 2006–2007 Ontario Graduate Scholarship ($15,000). 1998–2003 Term Dean’s Honour List four times (10th percentile), University of Waterloo. 1998–2003 Rated ‘Outstanding’ by co-op employers three times, University of Waterloo. (Requires written justification by workplace superviser.) JOURNAL PUBLICATIONS 4. B. Alipanahi* , A. Delong* , and B.J. Frey. Ascertaining sequence specificities using deep learning. Nature Biotechnology, in revision. (contributed equally* ) 3. M. Leung, A. Delong, B. Alipanahi, and B.J. Frey. Machine Learning in Genomics and Precision Medicine. IEEE Proceedings, submitted. 2. A. Delong, L. Gorelick, O. Veksler, and Y. Boykov. Minimizing Energies with Hierarchical Costs. International Journal of Computer Vision (IJCV), 100(1):38–58, October 2012. 1. A. Delong, A. Osokin, H. N. Isack, and Y. Boykov. Fast Approximate Energy Minimization with Label Costs. International Journal of Computer Vision (IJCV), 96(1):1–27, January 2012. REFEREED CONFERENCE PUBLICATIONS 10. L. Gorelick, Y. Boykov, O. Veksler, I. Ben Ayed, and A. Delong. Submodularization for Binary Pairwise Energies. In IEEE Computer Vision and Pattern Recognition (CVPR), June 2014. (oral) 9. A. Delong, O. Veksler, A. Osokin and Y. Boykov. Minimizing Sparse High-Order Energies by Submodular Vertex-Cover. In Neural Information Processing Systems (NIPS), December 2012. 8. A. Delong, O. Veksler, and Y. Boykov. Fast Fusion Moves for Multi-Model Estimation. In European Conference on Computer Vision (ECCV), October 2012. 7. L. Gorelick, F.R. Schmidt, Y. Boykov, A. Delong, and A. Ward. Segmentation with non-linear regional constraints via line-search cuts. In European Conference on Computer Vision (ECCV), October 2012. 6. L. Gorelick, A. Delong, O. Veksler, and Y. Boykov. Recursive MDL via Graph-Cuts: Application to Segmentation. In International Conference on Computer Vision (ICCV), September 2011. 5. A. Delong, L. Gorelick, F.R. Schmidt, O. Veksler, and Y. Boykov. Interactive Segmentation with Super-Labels. In Energy Min. Methods in Comp. Vis. (EMMCVPR), LNCS 6819, July 2011. (oral) 4. A. Delong, A. Osokin, H.N. Isack, and Y. Boykov. Fast Approximate Energy Minimization with Label Costs. In IEEE Computer Vision and Pattern Recognition (CVPR), June 2010. 3. A. Delong and Y. Boykov. Globally Optimal Segmentation of Multi-Region Objects. In International Conference on Computer Vision (ICCV), October 2009. (oral) 2. A. Delong and Y. Boykov. A Scalable Graph-Cut Algorithm for N-D Grids. In IEEE Computer Vision and Pattern Recognition (CVPR), June 2008. 1. Y. Boykov, V. Kolmogorov, D. Cremers, and A. Delong. An Integral Solution to Surface Evolution PDEs via Geo-Cuts. In European Conf. on Computer Vision (ECCV), LNCS 3953, May 2006. REFEREED WORKSHOP PUBLICATIONS 1. A. Delong, B. Alipanahi, and B.J. Frey. A Convolutional Model of RNA-Binding Proteins. In Neural Information Processing Systems (NIPS) Workshop on Machine Learning in Computational Biology, December 2013. (oral) PATENTS 2. H.Y. Xiong, A. Delong, and B.J. Frey. “Improving ensemble training by adjusting the variance across predictors”. United States Provisional Patent, 2015. 1. A. Delong, Y. Boykov, D. Yu. “Region Based Push-Relabel Algorithm for Efficient Computation of Maximum Flow”. United States Patent #7,844,113, 2010. INSTRUCTOR EXPERIENCE • CS 3388: Introduction to Computer Graphics (Fall 2011) – Authored complete set of course notes, assignments, exams. – Rated 6.3/7 overall teaching effectiveness by 34 enrolled students. (Previous course: 4.5/7.) • CS 1037: Fundamentals of Computer Science II (Fall 2010) – Authored complete set of course notes, assignments, exams, weekly labs. – Rated 6.4/7 overall teaching effectiveness by 93 enrolled students. (Previous course: 5.5/7.) ADDITIONAL TEACHING EXPERIENCE • Teaching Assistant, Analysis of Algorithms (Winter 2010) – Nominated by students for university-wide Teaching Assistant award. • Teaching Assistant, Fundamentals of Computer Science II (Fall 2009) • Course Developer, Fundamentals of Computer Science II (Fall 2008) – Revised lecture notes; taught many lectures/labs; developed electronic grading tools. • Teaching Assistant, Object-Oriented Design and Analysis (Winter 2008) – Recipient of 2008 Faculty of Science Graduate Student Teaching Award. • Teaching Assistant, Fundamentals of Computer Science II (Fall 2007) • Teaching Assistant, Fundamentals of Computer Science II (Fall 2006) – Nominated by students for university-wide Teaching Assistant award. • Teaching Assistant Training Program (TATP), University of Western Ontario (Fall 2006) WORK EXPERIENCE 2012–2015 University of Toronto, Postdoctoral Fellow (Brendan J. Frey) 2011 University of Western Ontario, Postdoctoral Fellow (Yuri Boykov & Olga Veksler) 2005 Siemens Corporate Research, Intern, Imaging and Visualisation (Princeton, NJ) – Explored designs for GPU max-flow algorithms for 3D volume segmentation. 2003–2004 Alias Systems, Software Engineer, Real-Time Architecture (Toronto, ON) – Studied rendering, networking, and physics in modern game engine architectures. 2000–2002 Alias|Wavefront, Co-op Software Engineer, Maya Development (Toronto, ON) 1998–1999 Pinpoint Information Systems, Co-op Software Engineer (Burlington, ON) INVITED TALKS 11. Skolkovo Institute of Science and Technology (Skoltech), Russia, May 2015. 10. Massachusetts Institute of Technology (MIT), USA, August 2013. 9. University of Toronto, Canada, September 2011. 8. Technical University of Munich (TUM), Germany, July 2011. 7. Lund University, Sweden, July 2011. 6. University of Heidelberg, Germany, July 2011. 5. École Supérieure d’Ingénieurs en Électronique et Électrotechnique (ESIEE), France, June 2011. 4. University of Oxford, United Kingdom, June 2011. 3. École Normal Supérieure (ENS), France, June 2011. 2. University of Toronto, Canada, December 2010. 1. Lund University, Sweden, May 2009. REFEREEING • Journals: – Nature Medicine – IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) – International Journal of Computer Vision (IJCV) – Transactions on Image Processing (TIP) – SIAM Journal on Imaging Sciences (SIIMS) – Image and Vision Computing (IMAVIS) – Journal of Electronic Imaging (JEI) – Pattern Recognition (PR) • Conference program committees: – Uncertainty in Artificial Intelligence (UAI) 2015 – International Conference on Computer Vision (ICCV) 2007–2011 – Computer Vision and Pattern Recognition (CVPR) 2007–2013 – European Conference on Computer Vision (ECCV) 2008–2014 – Medical Imaging and Computer Assisted Intervention (MICCAI) 2012 – Energy Min. Methods in Comp. Vision and Patt. Rec. (EMMCVPR) 2009–2013 – Genetic and Evolutionary Computation Conference (GECCO) 2013 – Eurographics (EG) 2007 • Workshop program committees: – Deep Learning Workshop at ICML 2015 NON-REFEREED CONTRIBUTIONS 6. Automatic Quantifier Elimination Proves a Key Result in Submodular Function Minimization. CAIMS Annual Meeting hosted by the Fields Institute, Symposium on Applications of computer algebra in applied and industrial mathematics, June 2012. (presentation) 5. Automatic Quantifier Elimination Proves a Key Result in Submodular Function Minimization. SIAM Conference on Discrete Mathematics, Symposium on interactions between computer algebra and discrete mathematics, June 2012. (presentation) 4. Database of max-flow problem instances in computer vision. http://vision.csd.uwo.ca/data/maxflow/ (cited by papers in CVPR and outside vision, e.g. by Andrew V. Goldberg and Dorit S. Hochbaum) 3. N-View Surface Feature Detection with AdaBoost. University of Western Ontario Research in Computer Science, April 2008. (presentation) 2. A Scalable Graph-Cut Algorithm for N-D Grids. University of Western Ontario Research in Computer Science, April 2008. (poster) 1. A Scalable Graph-Cut Algorithm for N-D Grids. Graph Cuts and Related Discrete or Continuous Optimization Problems, Institute for Pure and Applied Mathematics, UCLA, Feb 2008. (poster) ORGANIZATIONAL ACTIVITIES • • • • • • Organized weekly group seminars for Frey lab (2014–). Head organizer of departmental research conference (UWORCS 2009). Co-organizer of departmental research conference (UWORCS 2010–2012). Organized weekly seminars for computer vision research group (2007–2010). Created and maintained research group web server (2007–2012). http://vision.csd.uwo.ca/ Outreach to a local high school, promoting computer science and post-secondary education. LOCAL COMMITTEES • • • • • Society of Graduate Students (SOGS) councilor, University of Western Ontario (2006–2010). SOGS Health Plan Committee member, University of Western Ontario (2009–2010). Graduate Executive Committee student member, University of Western Ontario (2008–2009). Appointments Committee student member, University of Western Ontario (2007–2008). Executive member of UWO Computer Science Grad Student Council (CSGSC). COMPUTER SKILLS Applications: Visual Studio, GCC/G++, Matlab, VTune, Maya, Adobe Illustrator, Adobe Photoshop, LAMP/MediaWiki admin. Languages: C99/C++11, Python 2.7, CUDA, Matlab, x86 assembler, GLSL, Javascript, PHP, Java, C#, SQL, MEL, LaTeX. APIs: Numpy, Matplotlib, Matlab MEX, cuBLAS/cuRAND/cuDNN, OpenGL 3.0, DirectX 9, Win32/GDI, Maya Plugin API. REFERENCES available upon request March 26, 2015