CIS Colloquium, Mar 28, 2014, 11:00AM – 12:00PM, Wachman 1015D A Big 3D Data Approach to Visual Scene Understanding Jianxiong Xiao , Princeton University Abstract : On your one-minute walk from the coffee machine to your desk each morning, you pass by dozens of scenes — a kitchen, an elevator, your office — and you […]
Read More »CIS Colloquium, Mar 06, 2014, 11:00AM – 12:00PM, Wachman 1015D Large-Scale Structured Sparse Learning Jieping Ye , Arizona State University Abstract : Sparse methods have been shown to be a versatile and powerful tool for biomedical data analysis. In this talk, we consider sparse methods for (1) variable selection where the structure over the features […]
Read More »CIS Colloquium, Feb 19, 2014, 11:00AM – 12:00PM, Wachman 1015D Sampling Massive Datasets in the Internet Nicholas Duffield, Rutgers University Abstract : Massive graph datasets are used operationally by providers of Internet, social network and search services. Sampling can reduce storage requirements as well as query execution times, while prolonging the useful life of the […]
Read More »CIS Colloquium, Jan 28, 2014, 11:00AM – 12:00PM, Wachman 1015D Machine Learning in Evidence – Based Medicine: Making Better use of Domain Experts Byron Wallace, Brown University Abstract : An unprecedented volume of biomedical evidence is being published today. Indeed, PubMed (roughly, a search engine for biomedical literature) now indexes more than 600,000 publications describing […]
Read More »CIS Colloquium, Nov 22, 2013, 02:00PM – 03:00PM, Wachman 1015D Probabilistic Topic Models of Text and Users David Blei, Princeton University Abstract : Probabilistic topic models provide a suite of tools for analyzing large document collections. Topic modeling algorithms discover the latent themes that underlie the documents and identify how each document exhibits those themes. […]
Read More »CIS Distinguished Lecture Series, Nov 20, 2013, 11:00AM – 12:00PM, Barton 108A Computer Perception with Deep Learning Yann LeCun , School of Computer Science University of Massachusetts, Amherst Abstract : Pattern recognition tasks, particularly perceptual tasks such as vision and audition, require the extraction of good internal representations of the data prior to classification. Designing […]
Read More »CIS Colloquium, Nov 15, 2013, 11:00AM – 12:00PM, Wachman 1015D Information-Driven Imaging for Appearance Capture and Recognition Jinwei Gu, Rochester Institute of Technology Abstract : One of the fundamental problems in imaging, computer vision, and visualization is the study of appearance. Despite of tremendous success, the popular data – driven approach is soon becoming incompetent […]
Read More »CIS Colloquium, Nov 13, 2013, 11:00AM – 12:00PM, Wachman 447 Modeling Objects, Concepts, Aesthetics and Emotions in Images James Z. Wang, The Pennsylvania State University Abstract : Automated annotation of digital pictures has been a highly challenging open research problem. In our work, by advancing statistical modeling and optimization techniques, we can train computers about […]
Read More »CIS Distinguished Lecture Series, Nov 01, 2013, 11:00AM – 12:00PM, Tech Center 111 Using Process Modeling and Analysis Techniques to Reduce Errors in Healthcare Lori A. Clarke , School of Computer Science University of Massachusetts, Amherst Abstract : As has been widely reported in the news lately, healthcare errors are a major cause of death […]
Read More »CIS Colloquium, May 07, 2013, 11:00AM – 12:00PM, Wachman 1015D A Scalable Systems Approach to Big Data Challenges Hong Jiang, National Science Foundation University of Nebraska – Lincoln Abstract : Everyday 2.5 quintillion (2.5×1018, or 2.5 million trillion) bytes of data are created by people. This data comes from everywhere: from traditional scientific computing and […]
Read More »CIS Colloquium, May 03, 2013, 11:00AM – 12:00PM, Wachman 447 Designing Informatics Tools with Human Factors in Mind Rupananda Misra , Art Institute of Philadelphia Abstract : A number of Informatics tools have proliferated in the last few years in the domains such as healthcare and online education including educational games – which means, there […]
Read More »CIS Colloquium, May 01, 2013, 11:00AM – 12:00PM, Wachman 447 Analyzing the World’s Languages Ryan McDonald, Google Abstract : Building computer programs that can analyze and distill human language embodies the field of natural language processing (NLP). In the past decade, a number of significant achievements have been made due to the availability of large […]
Read More »CIS Colloquium, Apr 17, 2013, 11:00AM – 12:00PM, Wachman 447 Graph Estimation Han Liu, Princeton University Abstract : The graphical model has proven to be a useful abstraction in statistics and machine learning. The starting point is the graph of a distribution. While often the graph is assumed given, we are interested in estimating the […]
Read More »Vision Journal Club, Apr 17, 2013, 03:00PM – 04:00PM, Wachman 1015D High Resolution Registration, Tracking, and Synthesis of 3D Deforming Objects Dr Meizhu Liu, Siemens Corporate Research Abstract : A robust divergence measure is fundamental in both theoretical and practical applications. Dr. Liu proposed a new robust class of divergences, termed by total Bregman divergence […]
Read More »CIS Colloquium, Apr 16, 2013, 11:00AM – 12:00PM, Wachman 1015D Machine Learning with Big Linked Data Jun (Luke) Huan, University of Kansas Abstract : Graphs are widely used modeling tools that capture objects and their relation (links). Graph modeled data are found in diverse application areas including bioinformatics, cheminformatics, social networks, wireless sensor networks among […]
Read More »CIS Colloquium, Apr 03, 2013, 11:00AM – 12:00PM, Wachman 1015D Complex Contagion and The Weakness of Long Ties in Social Networks: Revisited Jie Gao, Stony Brook University Abstract : It is a common understanding that the diffusion of disease and information is fast in social networks, due to the existence of weak ties and the […]
Read More »CIS Colloquium, Feb 28, 2013, 11:00AM – 12:00PM, Wachman 1015D Unsupervised and Weakly-Supervised Object Discovery, Segmentation, Classification, and Clustering Zhuowen Tu , University of California, Los Angeles Abstract : It the first part of the talk, I will present two recent papers (CVPR 2012) along the line of weakly supervised learning for unsupervised object discovery […]
Read More »CIS Colloquium, Feb 27, 2013, 11:00AM – 12:00PM, Wachman 447 Data Exploration and Privacy Preservation Over Hidden Web Databases Nan Zhang , Comprehensive Cancer Center/Electrical and Computer Engineering, University of Alabama Abstract : A large number of online databases are hidden in the “deep web” and only accessible through restrictive search or browsing web interfaces. […]
Read More »CIS Colloquium, Feb 20, 2013, 11:00AM – 12:00PM, Wachman 447 Novel Methods in Medical Image Computing Allen Tannenbaum , Comprehensive Cancer Center/Electrical and Computer Engineering, University of Alabama Abstract : In this talk, we will describe some novel approaches to medical image computing, including segmentation and registration. Segmentation is the process of extracting key features […]
Read More »CIS Colloquium, Feb 13, 2013, 11:00AM – 12:00PM, Wachman 447 Data Driven Analytics for Personalized Healthcare Jianying Hu , IBM T. J. Watson Research Center Abstract : The concept of Learning Health Systems (LHS) is gaining momentum as more and more electronic healthcare data has become increasingly accessible. The idea is to enable learning from […]
Read More »CIS Colloquium, Feb 08, 2013, 11:00AM – 12:00PM, Wachman 447 The Roles of Uncertainty in 3D Reconstruction Philippos Mordohai , Stevens Institute of Technology Abstract : 3D reconstruction from two or more images is one of the most well-studied problems in computer vision. Due to the inverse nature of the problem, the reconstructed models typically […]
Read More »CIS Distinguished Lecture Series, Feb 06, 2013, 11:00AM – 12:00PM, Wachman 447 Computational Methods in the Digital Humanities Peter Melville Logan , Temple University Abstract : Computation is fast becoming as vital in humanistic scholarship as it already is in fields like plasma physics and climate modeling. “Digital Humanities” is a term that encapsulates all […]
Read More »CIS Colloquium, Feb 04, 2013, 11:00AM – 12:00PM, Tech Center 111 Intrinsically Disordered Proteins A. Keith Dunker , Indiana University School of Medicine Abstract : The standard view is that each protein’s amino acid sequence provides the information for it to fold into a specific 3D structure, and the active site formed within this structure […]
Read More »CIS Distinguished Lecture Series, Jan 30, 2013, 11:00AM – 12:00PM, Wachman 447 Sparse Linearized Iterative Coherence Estimation (SLICE) and Risk Assessment In Image Analysis and Systems Verification Robert Bonneau , Air Force Office of Scientific Research (AFOSR) Abstract : Many methods that form manifold learning have been proposed recently to accurately embed some high dimensional […]
Read More »CIS Colloquium, Dec 05, 2012, 11:15AM – 12:15PM, Wachman 447 Sparse Subspace Clustering Rene Vidal , Johns Hopkins University Abstract: In the era of data deluge, the development of methods for discovering structure in high-dimensional data is becoming increasingly important. Traditional approaches often assume that the data is sampled from a single low-dimensional manifold. However, […]
Read More »CIS Distinguished Lecture Series, Nov 29, 2012, 11:00AM – 12:00PM, Tech Center 111 Swarms: First Class Citizens in the Future Internet Don Towsley, University of Massachusetts Abstract : The current Internet consists of tens of thousand different interconnected autonomous networks. It was designed to support large populations of point-to-point content transfers. This introduces considerable headaches […]
Read More »CIS Colloquium, Nov 21, 2012, 11:00AM – 12:00PM, Wachman 447 Ranking, Retrieval and Recommendation using Supervised Embedding Models Jason Weston, Google, NY, USA Abstract: We present a class of latent embedding models that are discriminatively trained to map from the content in a query document or document-document pair to a ranking score. Like latent semantic […]
Read More »CIS Distinguished Lecture Series, Nov 14, 2012, 11:00AM – 12:00PM, Tech Center 111 Biological Evolution as a Form of Learning Leslie Valiant , Harvard University Abstract : Living organisms function according to protein circuits. Darwin’s theory of evolution suggests that these circuits have evolved through variation guided by natural selection. However, the question of which […]
Read More »CIS Colloquium, Nov 07, 2012, 11:15AM – 12:15PM, Wachman 447 Sparse Methods for Medical Image Analysis and Reconstruction Dimitris Metaxas, Rutgers University Abstract: Recent advances in sparse methods have resulted in efficient and more accurate solutions to a variety of medical imaging methods. In this talk we will present a new learning formulation called structured […]
Read More »CIS Distinguished Lecture Series, Nov 05, 2012, 11:00AM – 12:00PM, Tech Center 111 Explaining AdaBoost Robert Schapire, Princeton University Abstract : Boosting is an approach to machine learning based on the idea of creating a highly accurate prediction rule by combining many relatively weak and inaccurate rules. AdaBoost, the first practical boosting algorithm, has enjoyed […]
Read More »