Presentations

**(*) = Invited Talks**

Performance Portability of an SpMV Kernel Across Scientific Computing and Data Science Applications, Stephen L. Olivier, Nathan D. Ellingwood, Jonathan Berry, Daniel M. Dunlavy, Chesapeake Large-Scale Analytics Conference (CLSAC’21), ONLINE, October 2021.

Performance Portability of an SpMV Kernel Across Scientific Computing and Data Science Applications, Stephen L. Olivier, Nathan D. Ellingwood, Jonathan Berry, Daniel M. Dunlavy, IEEE High Performance Extreme Computing Conference (HPEC’21), ONLINE, September 2021.

Using Computation Effectively for Scalable Poisson Tensor Factorization: Comparing Methods Beyond Computational Efficiency, Jeremy Myers, Daniel M. Dunlavy, IEEE High Performance Extreme Computing Conference (HPEC’21), ONLINE, September 2021.

Performance-Portable Sparse Tensor Decomposition Kernels on Emerging Parallel Architectures, S. Isaac Geronimo Anderson, Keita Teranishi, Daniel M. Dunlavy, Jee Choi, IEEE High Performance Extreme Computing Conference (HPEC’21), ONLINE, September 2021.

Questa High Performance Data Analytics, Daniel M. Dunlavy, Jonathan Berry, Sandia Computing and Information Sciences External Review Board, ONLINE, March 2021.

SparTen: Leveraging Kokkos for On-node Parallelism in a Second Order Method for Fitting Canonical Polyadic Tensor Models to Poisson Data, Keita Teranishi, Daniel M. Dunlavy, Jeremy Myers, Richard F. Barrett, IEEE High Performance Extreme Computing Conference (HPEC’20), ONLINE, September 2020.

Parameter Sensitivity Analysis of the SparTen High Performance Sparse Tensor Decomposition Software, Jeremy Myers, Daniel M. Dunlavy, Keita Teranishi, David S Hollman, IEEE High Performance Extreme Computing Conference (HPEC’20), ONLINE, September 2020.

Multimodal Deep Learning for Flaw Detection in Software Programs, Scott Heidbrink, Kathryn N. Rodhouse, Daniel M. Dunlavy, Machine Learning and Deep Learning Workshop, ONLINE, August 2020.

Topic Modeling with Natural Language Processing for Identification of Nuclear Proliferation-Relevant Scientific and Technical Publications, Jonathan Bisila, Daniel M. Dunlavy, Zoe N. Gastelum, Craig D. Ulmer, Machine Learning and Deep Learning Workshop, ONLINE, August 2020.

(*) Tensor Decompositions for Analyzing Multi-Way Data, Daniel M. Dunlavy, CSRI Summer Seminar Series, Albuquerque, NM, July 2020.

Performance and Parallelization of CP-Alternate Poisson Regression Sparse Tensor Decomposition, Keita Teranishi, D.S. Hollman, Jeremy Myers, Daniel M. Dunlavy, SIAM Conference on Parallel Processing for Scientific Computing, Seattle, WA, February 2020.

Detecting Flaws in Software Programs using Multimodal Deep Learning, Daniel M. Dunlavy, Kathryn N. Rodhouse, Scott Heidbrink, Conference on Data Analysis, Santa Fe, NM, February 2020.

Natural Language Processing for Topic Identification supporting Document Search and Identification for Nuclear Proliferation Detection, Daniel M. Dunlavy, Jonathan Bisila, Zoe N. Gastelum, Conference on Data Analysis, Santa Fe, NM, February 2020.

Performance and Parallelization of CP-Alternate Poisson Regression Sparse Tensor Decomposition, Keita Teranishi, Richard F. Barrett, D.S. Hollman, Jeremy Myers, Daniel M. Dunlavy, AI and Tensor Factorizations for Physical, Chemical and Biological Systems, Santa Fe, NM, September 2019.

Applying Natural Language Processing (NLP) to RF Signal Patterns, Wesley Alexander Brooks, Daniel M. Dunlavy, Jennifer Galasso, Christopher M. Howerter, Nicholas Jacobs, Christine F. Lai, Machine Learning and Deep Learning Workshop, Albuquerque, NM, August 2019.

Development of Parallel Sparse CP-APR Tensor Decomposition Solvers, Keita Teranishi, Richard F. Barrett, D.S. Hollman, Daniel M. Dunlavy, Kokkos User Group Meeting, Albuquerque, NM, April 2019.

Performance Portable Parallel Sparse CP-APR Tensor Decompositions, Keita Teranishi, Richard F. Barrett, Christopher J. Forster, Tamara G. Kolda, Daniel M. Dunlavy, SIAM Conference on Computational Science and Engineering, Spokane, WA, February 2019.

Exploiting Scientific Software to Solve Problems in Data Analytics, Karen D. Devine, Erik Boman, Tamara G. Kolda, Michael Wolf, Daniel M. Dunlavy, IPAM Workshop III: HPC for Computationally and Data-Intensive Problems, Los Angeles, CA, November 2018.

Static Source Code Analysis Using Neural Networks, Gary J. Saavedra, Daniel M. Dunlavy, Machine Learning and Deep Learning Workshop, Albuquerque, NM, July 2018.

A Performance Portable Parallel CP-APR Decomposition, Keita Teranishi, Richard F. Barrett, Christopher J. Forster, Tamara G. Kolda, Daniel M. Dunlavy, Sparse Days, Toulouse, France, September 2018.

Scalable Tensor Factorizations on Multiple Architectures, Daniel M. Dunlavy, Christopher J. Forster, Keita Teranishi, Tamara G. Kolda, Three-way Methods in Chemistry and Psychology (TRICAP), Angel Fire, NM, June 2018.

Analysis of Performance and Portability of Sparse Tensor Decompositions on CPU/MIC/GPU Architectures using Kokkos, Christopher J. Forster, Keita Teranishi, Tamara G. Kolda, Daniel M. Dunlavy, NVIDIA GPU Technology Conference, San Jose, CA, March 2018.

Event Detection in Multivariate Scientific Simulations using Feature Anomaly Metrics, Aditya Konduri, Hemanth Kolla, Julia Ling, W. Philip Kegelmeyer, Daniel M. Dunlavy, Timothy M. Shead, Warren L. Davis IV, SIAM Conference on Parallel Computing, Tokyo, Japan, March 2018.

Analysis of Performance and Portability of Sparse Tensor Decompositions on CPU/MIC/GPU Architectures, Christopher J. Forster, Keita Teranishi, Tamara G. Kolda, Daniel M. Dunlavy, SIAM Conference on Parallel Computing, Tokyo, Japan, March 2018.

TriData: High Performance Linear Algebra-Based Data Analytics, Michael Wolf, Richard B. Lehoucq, Daniel Bourgeois, Alicia M. Klinvex, Daniel M. Dunlavy, SIAM Conference on Parallel Computing, Tokyo, Japan, March 2018.

Portability and Scalability of Sparse Tensor Decompositions on CPU/MIC/GPU Architectures, Keita Teranishi, Christopher J. Forster, Tamara G. Kolda, Daniel M. Dunlavy, Advanced Supercomputing Environment (ASE) Seminar, Tokyo, Japan, August 2017.

Portability and Scalability of Sparse Tensor Decompositions on CPU/MIC/GPU Architectures, Christopher J. Forster, Keita Teranishi, Greg E. Mackey, Tamara G. Kolda, Daniel M. Dunlavy, SIAM Annual Meeting, Pittsburgh, PA, July 2017.

Portability and Scalability of Sparse Tensor Decompositions on CPU/MIC/GPU
Architectures, Daniel M. Dunlavy, HPC Analytics Workshop, Hanover, MD, June 2017.

A Computational Spectral Graph Theory Tutorial, Richard B. Lehoucq, Erik G. Boman, Karen D. Devine, Jonathan W. Berry, Michael Wolf, Van Henson, Geoff Sanders, Daniel M. Dunlavy, Householder Symposium XX, Blacksburg, VA, June 2017.

TriData: Applying Trilinos to Data Analysis, Michael Wolf, Alicia M. Klinvex, Daniel M. Dunlavy, Trilinos User-Developer Group Meeting, Albuquerque, NM, October 2016.

Hypergraph Exploitation for Data Sciences, Michael Wolf, Alicia M. Klinvex, Daniel M. Dunlavy, Graph Exploitation Symposium, Dedham, MA, May 2016.

Clustering Network Data through Effective Use of Eigensolvers and Hypergraph Models, Alicia M. Klinvex, Michael Wolf, Daniel M. Dunlavy, Copper Mountain Conference on Iterative Methods, Copper, CO, March 2016.

(*) Clustering Network Data using Graphs, Hypergraphs, and Tensors, Daniel M. Dunlavy, Statistical and Computational Challenges in Networks and Cybersecurity, Montreal, Canada, May 2015.

(*) Data Analysis using Tensor Factorizations, Daniel M. Dunlavy, CSRI Summer Seminar Series, Albuquerque, NM, July 2013.

Community Detection: A Bayesian Approach and the Challenge of Evaluation, Jonathan W. Berry, Daniel M. Dunlavy, Cynthia A. Phillips, David G. Robinson, Jiqiang Guo, Daniel Nordman, Alyson Wilson, Structure, Statistical Inference, and Dynamics Networks: From Graphs to Rich Data, Santa Fe, NM, May 2013.

Community Detection: A Bayesian Approach and the Challenge of Evaluation, Jonathan W. Berry, Daniel M. Dunlavy, Cynthia A. Phillips, David G. Robinson, Jiqiang Guo, Daniel Nordman, Alyson Wilson, Graph Exploitation Workshop, Dedham, MA, April 2012.

Parallel Bayesian Methods for Community Detection, Jonathan W. Berry, Daniel M. Dunlavy, Cynthia A. Phillips, David G. Robinson, Jiqiang Guo, Daniel Nordman, Alyson Wilson, SIAM Conference on Parallel Computing, Savannah, GA, February 2012.

(*) Text Analysis, Factor Modeling and Machine Learning, Daniel M. Dunlavy, EECS Department Seminar, Northwestern University, Evanston, IL, November 2011.

Scalable Tensor Factorizations with Incomplete Data, Tamara G. Kolda, Evrim Acar Ataman, Daniel M. Dunlavy, Morten Mørup, SIAM Annual Meeting, Pittsburgh, PA, July 2010.

Compressively Sensed Complex Networks, Jaideepp Ray, Ali Pinar, Daniel M. Dunlavy, SIAM Annual Meeting, Pittsburgh, PA, July 2010.

ParaText: Scalable Text Analysis and Visualization, Daniel M. Dunlavy, Eric T. Stanton, Timothy M. Shead, SIAM Annual Meeting, Pittsburgh, PA, July 2010.

Link Prediction on Evolving Graphs using Matrix and Tensor Factorizations, Tamara G. Kolda, Evrim Acar Ataman, Daniel M. Dunlavy, BIT 50 – Trends in Numerical Computing, Lund, Sweden, June 2010.

ParaText: Scalable Text Analysis and Visualization, Daniel M. Dunlavy, Eric T. Stanton, Timothy M. Shead, SIAM Conference on Parallel Processing for Scientific Computing, Seattle, WA, February 2010.

Scalable Tensor Factorizations with Incomplete Data, Tamara G. Kolda, Evrim Acar Ataman, Daniel M. Dunlavy, Morten Mørup, Sparse Random Structures: Analysis and Computation, Banff, AB, Canada, January 2010.

Coupled Tensor and Matrix Factorizations, Tamara G. Kolda, Evrim Acar Ataman, Daniel M. Dunlavy, SIAM Conference on Applied Linear Algebra, Monterey Bay-Seaside, CA, October 2009.

Scalable Tensor Factorizations with Missing Data, Evrim Acar Ataman, Tamara G. Kolda, Daniel M. Dunlavy, SIAM Conference on Applied Linear Algebra, Monterey Bay-Seaside, CA, October 2009.

Persistent Homology for Parameter Sensitivity in Large-scale Text-analysis (Informatics) Graphs, Daniel M. Dunlavy, CSRI Workshop on Combinatorial Algebraic Topology (CAT), Santa Fe, NM, August 2009.

ISE Informatics: Statistical Analysis of Heterogeneous Data for Test Quality Assurance, Brett W. Bader, Mark Van Benthem, Shawn Martin, Daniel M. Dunlavy, Compatibility, Aging & Stockpile Stewardship Conference, Albuquerque, NM, September 2009.

Scalable Solutions for Processing and Searching Very Large Document Collections, Daniel M. Dunlavy, Eric T. Stanton, Timothy M. Shead, NNSA LDRD Symposium, Washington, D.C., August 2009.

An Optimization Approach for Fitting Canonical Tensor Decompositions, Daniel M. Dunlavy, Evrim Acar Ataman, Tamara G. Kolda, SIAM Annual Meeting, Denver, CO, July 2009.

An Optimization Approach for Fitting a CANDECOMP/PARAFAC Model with Applications in Social Network Analysis, Evrim Acar Ataman, Tamara G. Kolda, Daniel M. Dunlavy, Three-way Methods in Chemistry and Psychology (TRICAP), Vall de Nuria, Spain, June 2009.

The Canonical Tensor Decomposition and Its Applications to Data Analysis, Evrim Acar Ataman, Tamara G. Kolda, Daniel M. Dunlavy, Linear Algebra and Optimization Seminar, Palo Alto, CA, May 2009.

Link Prediction on Evolving Data using Tensor Factorizations, Evrim Acar Ataman, Tamara G. Kolda, Daniel M. Dunlavy, SIAM Conference on Computational Science and Engineering, Miami, FL, March 2009.

Coupled Matrix Factorizations Using Optimization, Daniel M. Dunlavy, Evrim Acar Ataman, Tamara G. Kolda, SIAM Conference on Computation Science and Engineering, Miami, FL, March 2009.

ParaText: Leveraging Scalable Scientific Computing Capabilities for Large-Scale Text Analysis and Visualization, Daniel M. Dunlavy, Eric T. Stanton, Timothy M. Shead, SIAM Conference on Computation Science and Engineering, Miami, FL, March 2009.

CPOPT: Optimization for Fitting CANDECOMP/PARAFAC Models, Tamara G. Kolda, Evrim Acar Ataman, Daniel M. Dunlavy, Computational Algebraic Statistics, Theories and Applications (CASTA2008), Kyoto, Japan, December 2008.

CPOPT: Optimization for Fitting CANDECOMP/PARAFAC Models, Tamara G. Kolda, Evrim Acar Ataman, Daniel M. Dunlavy, Multi-Manifold Data Modeling and Applications, Minneapolis, MN, October 2008.

Multilinear Algebra for Analyzing Data with Multiple Linkages, Daniel M. Dunlavy, Tamara G. Kolda, W. Philip Kegelmeyer, DOE Applied Mathematics Principal Investigators Meeting, Argonne, IL, October 2008.

(*) Text Analysis: Methods for Searching, Organizing, Labeling and Summarizing Document Collections, Daniel M. Dunlavy, CSRI Student Seminar Series, Albuquerque, NM, July 2008.

Heterogeneous Ensemble Classification, Daniel M. Dunlavy, Sean A. Gilpin, Sandia Workshop on Data Mining and Data Analysis, Albuquerque, NM, July 2008.

Coupling Informatics Algorithm Development and Visual Analysis, Daniel M. Dunlavy, SIAM Annual Meeting, San Diego, CA, July 2008.

Tensor Decompositions for Analyzing Multi-link Graphs, Daniel M. Dunlavy, Tamara G. Kolda, W. Philip Kegelmeyer, SIAM Conference on Parallel Processing for Scientific Computing, Atlanta, GA, March 2008.

The Titan Project: Requirements for Data Mining & Analysis,, Daniel M. Dunlavy, Timothy M. Shead, Patricia J. Crossno, Eric T. Stanton, Trilinos User Group Meeting, Albuquerque, NM, November 2007.

(*) Multilinear Algebra for Analyzing Data with Multiple Linkages, Daniel M. Dunlavy, Tamara G. Kolda, W. Philip Kegelmeyer, Department of Mathematics Colloquium, Western Michigan University, Kalamazoo, MI, September 2007.

(*) Applying Mathematics: Solving Problems at a National Laboratory, Daniel M. Dunlavy, Pi Mu Epsilon Talk, Western Michigan University, Kalamazoo, MI, September 2007.

(*) Planning for a Successful Career at a National Laboratory, Daniel M. Dunlavy, Graduation Conference, University of Maryland, College Park, MD, April 2007.

(*) The Use of Homotopies and Continuation in Solving Challenging Problems in the Physical and Life Sciences, Daniel M. Dunlavy, Department of Mathematics and Computer Science Colloquium, Chicago State University, Chicago, IL, March 2007.

Formulations for Surrogate-Based Optimization with Data Fit and Multifidelity Models, Daniel M. Dunlavy, Michael S. Eldred, SIAM Conference on Computational Science and Engineering, Costa Mesa, CA, February 2007.

Periodic Orbits with 4D, Andrew G. Salinger, Eric T. Phipps, Daniel M. Dunlavy, Trilinos Users Group Meeting, Albuquerque, NM, November 2006.

Global Optimization: For Some Problems, There’s HOPE, Daniel M. Dunlavy, Dianne P. O’Leary, Ninth Copper Mountain Conference on Iterative Methods, Copper Mountain, CO, April 2006.

Preconditioners for the Space-Time Solution of Large-Scale PDE Applications, Daniel M. Dunlavy, Andrew G. Salinger, SIAM Conference on Parallel Processing, San Francisco, CA, February 2006.

(*) Global Optimization: For Some Problems, There’s HOPE, Daniel M. Dunlavy, Dianne P. O’Leary, Argonne National Laboratory, Argonne, IL, March 2005.

(*) Global Optimization: For Some Problems, There’s HOPE, Daniel M. Dunlavy, Dianne P. O’Leary, Sandia National Laboratories, Livermore, CA, March 2005.

A Homotopy Method for Predicting Low Energy Conformations of Proteins, Daniel M. Dunlavy, Dianne P. O’Leary, SIAM Conference on Computational Science and Engineering, Orlando, FL, February 2005.

A Homotopy Method for Finding Low Energy Conformations of Polypeptides, Daniel M. Dunlavy, SIAM Conference on the Life Sciences, Portland, OR, July 2004.

Structure Preserving Eigensolvers, Daniel M. Dunlavy, Niloufer Mackey, D. Steven Mackey, SIAM Applied Linear Algebra Meeting, Williamsburg, VA, July 2003.

(*) A Homotopy Method for Predicting the State of Minimal Energy for Chains of Charged Particles, Daniel M. Dunlavy, Spotlight on Graduate Research Winner’s Lecture, Department of Mathematics, University of Maryland, February 2003.

A Homotopy Method for Predicting the State of Minimal Energy for Chains of Charged Particles, Daniel M. Dunlavy, Graduate Research Interaction Day, University of Maryland, April 2002.

Mathematical Modeling in Industry: Notes from a Graduate Workshop, Daniel M. Dunlavy, Pi Mu Epsilon Colloquium Series, Western Michigan University, January 2001.