Ben Taskar
Associate Professor
Computer and Information Science (CIS)
Honors and Awards: NSF CAREER Award - 2011, Office of Naval Research Young Investigator - 2010, Sloan Research Fellowship - 2010, DARPA Computer Science Study Group - 2009
Research Expertise: Machine Learning | Computational Linguistics | Computer Vision
Ben's research is focused on machine learning. He is interested in the design and analysis of data-centric algorithms that enable computers to tackle intelligent tasks, from machine translation and information extraction in natural language processing to object and action recognition in computer vision. Blending ideas from probability and statistics, discrete and continuous optimization, linguistic and perception theory, Ben's research aims to create drastically more accurate and robust systems that can sense, reason and act under real world uncertainty and ambiguity.
Member of:- General Robotics, Automation, Sensing and Perception (GRASP) Lab
- Institute for Research in Cognitive Science (IRCS)
- Penn Research in Machine Learning (PRiML)
Recent Publications 
- Determinantal point processes for machine learning, Kulesza, A. | Taskar, B., Foundations and Trends in Machine Learning, 2012
- Near-optimal MAP inference for determinantal point processes, Gillenwater, J. | Kulesza, A. | Taskar, B., Advances in Neural Information Processing Systems, 2012
- Shape-based object detection via boundary structure segmentation, Toshev, A. | Taskar, B. | Daniilidis, K., International Journal of Computer Vision, 2012
- Generative-discriminative basis learning for medical imaging, Batmanghelich, N.K. | Taskar, B. | Davatzikos, C., IEEE Transactions on Medical Imaging, 2012
- Regularized tensor factorization for multi-modality medical image classification., Batmanghelich, N. | Dong, A. | Taskar, B. | Davatzikos, C., Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention, 2011