Selected Presentations

Invited Talks at Conferences

• Estimating Optimal Dynamic Treatment Regimes with Shared Decision Rules across Stages: An Extension of Q-learning. Causal Inference in Health Research Workshop, Centre de Recherches Mathematiques, Montreal, Canada. May, 2011.

• Estimating Optimal Dynamic Treatment Regimes with Shared Decision Rules across Stages: An Extension of Q-learning. IISA Conference on Probability, Statistics, and Data Analysis, Raleigh, North Carolina. April, 2011.

• Estimating Optimal Dynamic Treatment Regimes with Shared Decision Rules across Stages: An Extension of Q-learning.Invited session on ``Recent Method Development on Reinforcement Learning and Personalized Medicine'' at the ENAR Spring Meetings of the International Biometric Society, Miami, Florida. March, 2011. 

• Estimating Optimal Dynamic Treatment Regimes with Shared Decision Rules across Stages: An Extension of Q-learning. WNAR-sponsored invited session on "Advances and Applications in Dynamic Treatment Regimes" at the Joint Statistical Meetings, Vancouver, Canada. August, 2010. [slides] 

• Inference for Non-regular Parameters in Optimal Dynamic Treatment Regimes. IMS-sponsored invited session on "Dynamic Treatment Regimes and Reinforcement Learning in Clinical Trials", ENAR Spring Meetings of the International Biometric Society, New Orleans, Louisiana. March, 2010. [slides]

• Inference for Non-regular Parameters in Optimal Dynamic Treatment Regimes. Special invited session on "Innovations of Statistics for Biology and Medicine" at the International Conference on Statistics, Probability, Operations Research, Computer Science and Allied Areas, Visakhapatnam, India. January, 2010. 

• Experimental Designs for Building and Refining CAM Interventions. Workshop on "Applying Principles from Complex Systems to Studying the Efficacy of CAM Therapies" at the Georgetown University School of Medicine, Washington, DC.October, 2007. 

• A Novel Approach for Developing Whole System Herbal Therapies. International Congress on Complementary Medicine Research, Munich, Germany. May, 2007.

• The Multi-phase Optimization Strategy: A Novel Way to Develop Multi-component Behavioral Interventions. ENAR Spring Meetings of the International Biometric Society, Atlanta, Georgia. March, 2007.

Invited Seminars

• Estimating Optimal Dynamic Treatment Regimes with Shared Decision Rules across Stages: An Extension of Q-learning. Causal Inference Working Group, Department of Biostatistics, Johns Hopkins University, Baltimore, Maryland. September, 2011.
• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. Department of Biostatistics & Epidemiology, University of Pennsylvania, Philadelphia, Pennsylvania. February, 2010.

• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. Department of Epidemiology, Biostatistics & Occupational Health, McGill University, Montreal, Canada. September, 2009.

• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. Department of Preventive Medicine, Northwestern University, Chicago. March, 2009. 

• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. Department of Biostatistics, Columbia University, New York. March, 2009. 

• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. Department of Statistics, Purdue University, West Lafayette, Indiana. March, 2009. 

• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. RAND Statistics Group, Santa Monica, California. March, 2009. 

• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. Department of Biostatistics, University of North Carolina, Chapel Hill, North Carolina. February, 2009. 

• Multistage Decision Policies via Q-learning: Bias Correction and Confidence Intervals. Bell Labs, Murray Hill, New Jersey. January, 2009. 

• Inference for Multistage Decision Policies via Regularized Q-Learning. Yahoo! Research Lab, Santa Clara, California. March, 2008.

• Design Strategies for Behavioral Intervention Research: A New Direction. The Methodology Center Brownbag Seminar Series, Pennsylvania State University, State College, Pennsylvania. March, 2005.

Contributed Talks

• Inference for Non-regular Parameters in Optimal Dynamic Treatment Regimes. ENAR Spring Meetings, San Antonio, Texas. March, 2009.

• Bias Correction and Confidence Intervals for Fitted Q-iteration. Neural Information Processing Systems (NIPS) workshop on "Model Uncertainty and Risk in Reinforcement Learning", Whistler, Canada. December, 2008. 

• Developing Multicomponent Interventions using Fractional Factorial Designs. Rackham Interdisciplinary Statistics Student Seminar Series, Ann Arbor, Michigan. July, 2008. 

• Inference for Dynamic Treatment Regimes via Q-Learning. Joint Statistical Meetings, Salt Lake City, Utah. July, 2007. 

• The Multi-phase Optimization Strategy: A New Way to Develop Multi-component Interventions. ENAR Spring Meetings of the International Biometric Society, Tampa, Florida. March, 2006. 

Posters

• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. Causal Inference in Statistics and the Quantitative Sciences Workshop, Banff, Canada. May, 2009. 

• Dealing with Non-regularity in Optimal Dynamic Treatment Regimes. Michigan Student Symposium for Interdisciplinary Statistical Sciences, Ann Arbor, Michigan. April, 2009.

• Estimation and Inference for Optimal Dynamic Treatment Regimes. Michigan Student Symposium for Interdisciplinary Statistical Sciences, Ann Arbor, Michigan. March, 2007.