Invited Talk (Posner Lecture)

Challenges for Machine Learning in Computational Sustainability

Thomas Dietterich

2012 Invited Talk (Posner Lecture)

Abstract

Research in computational sustainability seeks to develop and apply methods from computer science to the many challenges of managing the earth's ecosystems sustainably. Viewed as a control problem, ecosystem management is challenging for two reasons. First, we lack good models of the function and structure of the earth's ecosystems. Second, it is difficult to compute optimal management policies because ecosystems exhibit complex spatio-temporal interactions at multiple scales.

This talk will discuss some of the many challenges and opportunities for machine learning research in computational sustainability. These include sensor placement, data interpretation, model fitting, computing robust optimal policies, and finally executing those policies successfully. Examples will be discussed on current work and open problems in each of these problems.

All of these sustainability problems involve spatial modeling and optimization, and all of them can be usefully conceived in terms of facilitating or preventing flows along edges in spatial networks. For example, encouraging the recovery of endangered species involves creating a network of suitable habitat and encouraging spread along the edges of the network. Conversely, preventing the spread of diseases, invasive species, and pollutants involves preventing flow along edges of networks. Addressing these problems will require advances in several areas of machine learning and optimization.

Speaker

Thomas Dietterich

Tom Dietterich (AB Oberlin College 1977; MS University of Illinois 1979; PhD Stanford University 1984) is Professor and Director of Intelligent Systems Research at Oregon State University. Among his contributions to machine learning research are (a) the formalization of the multiple-instance problem, (b) the development of the error-correcting output coding method for multi-class prediction, (c) methods for ensemble learning, (d) the development of the MAXQ framework for hierarchical reinforcement learning, and (e) the application of gradient tree boosting to problems of structured prediction and latent variable models. Dietterich has pursued application-driven fundamental research in many areas including drug discovery, computer vision, computational sustainability, and intelligent user interfaces. Dietterich has served the machine learning community in a variety of roles including Executive Editor of the Machine Learning journal, co-founder of the Journal of Machine Learning Research, editor of the MIT Press Book Series on Adaptive Computation and Machine Learning, and editor of the Morgan-Claypool Synthesis series on Artificial Intelligence and Machine Learning. He was Program Co-Chair of AAAI-1990, Program Chair of NIPS-2000, and General Chair of NIPS-2001. He was first President of the International Machine Learning Society (the parent organization of ICML) and served a term on the NIPS Board of Trustees and the Council of AAAI.

Chat is not available.