Dr. Jayadev Acharya is interested in information theory, statistics, and machine learning. His research interests lie in designing efficient algorithms and understanding the fundamental limits of statistical and machine learning problems. He is interested in machine learning in medicine, for many reasons. The data is expensive, and high-dimensional with categorical (discrete) features. High dimensional categorical random variables are much less well understood than their continuous counterpart. It would be interesting to develop models, and understand them well enough to extract pertinent features using only a few medical records. Another aspect is privacy, which is a key constraint in addition to the others mentioned above, especially in medical applications. The trade-offs of privacy with the other resources would be an exciting avenue to explore.
Dr. Sumanta Basu is broadly interested in developing statistical machine learning methods for structure learning and prediction of complex, high-dimensional systems arising in biological and social sciences. He is currently working in two areas: (a) network modeling of high-dimensional time series; and (b) detecting high-order interactions in complex biological systems using randomized tree ensembles.
Dr. Florentina Buena's research interests include machine learning and empirical processes theory. She is particularly interested in model selection and averaging (aggregation) in a variety of high dimensional parametric, semi-parametric and nonparametric models. Her focus is on defining appropriate notations of sparsity and analyzing theoretically computationally efficient estimators tailored to sparse models, in problems where the number of parameters exceeds the sample size (" p > > n" problems). She has recently become interested in inference in high dimensional matrix models.
Morten H. Christiansen received his PhD in Cognitive Science from the University of Edinburgh in 1995. He is Professor in the Department of Psychology and Co-Director of the Cognitive Science Program at Cornell University as well Senior Scientist at the Haskins Labs, Professor of Child Language at the Interacting Minds Centre at Aarhus University and Professor in the Department of Language and Communication at the University of Southern Denmark. His research focuses on the interaction of biological and environmental constraints in the evolution, acquisition and processing of language. He employs a variety of methodologies, including computational modeling, corpus analyses, statistical learning, psycholinguistic experiments, and neuroimaging. Christiansen is the author of more than 175 scientiﬁc papers and has edited books on Connectionist Psycholinguistics, Language Evolution, Language Universals, and Cultural Evolution. His newest book Creating language: Integrating evolution, acquisition, and processingfrom MIT Press provides an overview of the work done in the Cognitive Neuroscience Lab.
Dr. Giles Hooker works on diagnostics and uncertainty quantification for machine learning. Neural networks, random forests and other methods tend to produce "black-box" models that produce predictions but little insight. He is interested in finding ways to quantify how stable those predictions are, and to then use this to ask questions about which inputs are important, how they change predictions and how they interact with other inputs. He has had successful collaborations on data from ecology, from medical records and immune processes.
The theme of Dr. Ning's research is to develop statistical methods and theory to quantify the uncertainty (confidence interval and hypothesis test) in modern data sets, which are characterized by high dimensionality, complexity and heterogeneity. I enjoy working at the interface of mathematical statistics, machine learning and stochastic optimization. I am also interested in applied projects in genomics, neuroscience, epidemiology and clinical trials.
Dr. Anthony Reeves supervises The Vision and Image Analysis group (VIA), which conducts research on automated computer methods for analyzing digital images especially with regards to accurate image measurements and with a primary focus on biomedical applications. A main research objective is the automatic detection and diagnosis of lung cancer from low-dose Computer Tomography (CT) scans and the computer aided diagnosis of diseases within the chest. An important outgrowth of the work has been the development of unique web-based clinical-study data-management system for the creation of large datasets that includes both clinical data and all medical images.
Dr. David Ruppert is the Andrew Schultz, Jr., Professor of Engineering, School of Operations Research and Information Engineering and Professor of Statistical Science, Cornell University. His current research focuses on astrostatistics, neuroscience, measurement error models, functional data analysis, semiparametric regression, and environmental statistics.
Dr. Mert Sabuncu's research interests include:
• Biomedical image analysis, with application focus in neurology/neuroscience
• Machine Learning, pattern recognition, multivariate statistics, Bayesian graphical models, approximate inference
• Data mining, applied to large-scale biomedical datasets, including genetics and imaging modalities
• Computational imaging genetics
• Image processing, computer vision
Dr. Karthik Sridharan is interested in design and analysis of machine learning algorithms with specific focus on developing learning algorithms with provable guarantees on performance. Of late he has been focusing on machine learning problems on graphs, specifically, online prediction algorithms on social networks and other forms of complex networks. The key challenges are to address the dynamic and ever changing nature of these prediction problems along with an eye for developing fast and efficient algorithms.
Dr. Martin T. Wells, Ph.D., is the Charles A. Alexander Professor of Statistical Sciences at Cornell-Ithaca. He is also a Professor of Social Statistics, Professor of Clinical Epidemiology and Health Services Research at Weill Medical School, an Elected Member of the Cornell Law School Faculty, as well as the Director of Research in the School of Industrial and Labor Relations. His research interests include statistical modeling for high-dimensional data, variable selection, multivariate analysis, Bayesian statistics, causal inference, and statistical methods for analyzing complex biological and legal data. He is also interested in statistical methods in machine learning.
Dr. Andrew G. Wilson develops flexible, interpretable, and scalable machine learning models, particularly for kernel learning and deep learning. He has expertise in probabilistic modelling, Gaussian processes, Bayesian nonparametrics, kernel methods, neural networks, scalable algorithms, and automatic machine learning. His work has been applied to time series, image, and video extrapolation, vision, geostatistics, gene expression, epidemiology, natural sound modelling, kernel discovery, Bayesian optimisation, econometrics, cognitive science, NMR spectroscopy, PET imaging, spatiotemporal statistics, and general relativity.
Dr. Haiyuan Yu performs research in the broad area of Network Systems Biology with both high-throughput experimental (see Vo et al., Cell 2016) and integrative computational (see Wang et al., Nature Biotechnology 2012) methodologies, aiming to understand gene functions and their relationships within complex molecular networks and how perturbations to such systems may lead to various human diseases. Using machine learning approaches, he is particularly interested in proteome-scale structural modeling and their applications in functional genomics studies, as well as predictions of functional impact of coding and non-coding mutations.
Qing Zhao's research interests include machine learning, sequential decision theory, statistical inference, and algorithmic theory with applications in infrastructure and communication networks, social economic networks, medicine, and computational biology. Her current research projects focus on sequential design of experiments for anomaly detection, risk-averse online learning and multiple comparisons problem with applications in clinical trials, combination therapy, and gene testing.