aprÃl, 2007
5 stars based on
71 reviews
Friday, March 8, at 2: We combine a scenario-based, standard-response optimization model with stochastic simulation to improve the efficiency of resource deployment for initial attack on wildland fires in three planning units in California. We use the California Fire Economics Simulator to predict the number of fires not contained within initial attack modeling limits.
Compared with the current deployment, the deployment obtained with optimization shifts resources from the planning unit with highest fire load to the planning unit with the highest standard response requirements leaving simulated containment success unchanged. This result suggests that, under the current budget and capacity constraints, a range of deployments may perform equally well in terms of fire containment.
When constraints on firefighting budget and station capacity are relaxed, the optimization produces deployments with greater containment success, suggesting that fire suppression effectiveness will be negatively impacted by declining budgets and improved by consolidating existing resources into fewer stations.
Friday, February 1, at 2: Identifying coherent sub-graphs in networks is important in many applications. In power systems, large systems are divided into areas and zones theo beijer exxonmobil fuels marketing aid in planning and control applications. But not every partitioning is equally good for all applications. In this talk we will discuss theo beijer exxonmobil fuels marketing hybrid method that combines a conventional graph partitioning algorithm with a genetic algorithm to partition a power network based on electrical distances, cluster sizes, the number of clusters, and cluster connectedness.
Clusters produced by this method can be used to identify buses with dynamically coherent voltage angles, without the need for dynamic simulation.
This also results in intra-zone transactions that have less impact on power flows outside of the zone, a property particularly useful for power system applications where ensuring deliverability is important, such as transmission planning or determination of synchronous reserve zones. Friday, November 30, at 2: Instance theo beijer exxonmobil fuels marketing for multi-instance theo beijer exxonmobil fuels marketing learning, with applications to object recognition in images and bird song.
In supervised learning problems concerning images and sounds, it is often natural to decompose a document bag into a collection of parts instancesand to associate each instance with a feature vector. It theo beijer exxonmobil fuels marketing also natural to associate documents with multiple labels, for example describing theo beijer exxonmobil fuels marketing objects in an image, or all species in an audio recording of bird song.
This structure motivates the multi-instance multi-label MIML framework, where the goal is to learn a classifier from a dataset consisting of bags of instances paired with sets of labels. In a MIML dataset, instances are not labeled directly; only bags are labeled.
Labeling at this course scale usually requires less human effort than labeling instances. Prior work on MIML has focused on predicting the label set for a previously unseen bag.
We instead consider the problem of predicting instance labels, while learning only from data theo beijer exxonmobil fuels marketing at the bag level. This problem is called instance annotation for MIML. We propose a regularized rank-loss objective designed for instance annotation, which can be instantiated with different aggregation models that link bag-level loss with instance-level predictions.
We consider aggregation models that can be factored as a linear function of one "support instance" per class, which are feature vectors summarizing a bag. Hence we name our proposed method rank-loss Theo beijer exxonmobil fuels marketing Instance Machines.
We propose two optimization methods for the rank-loss objective, which is non-convex. One is a heuristic that alternates between updating support instances and solving a convex problem in which the support instances are treated as constant. The other is to apply the constrained concave-convex procedure CCCPwhich solves a similar convex problem in each step.
Additionally, we suggest a method of extending the linear learning algorithm to non-linear classification without increasing asymptotic runtime.
Experiments show that the proposed methods achieve better theo beijer exxonmobil fuels marketing than recent methods based on other loss functions. Friday, November 2, at 2: This viral disease theo beijer exxonmobil fuels marketing yield, delays fruit ripening, and affects wine theo beijer exxonmobil fuels marketing.
The disease ecology is still under study and the spatial-dynamics of the spread process remains poorly understood. Moreover, little is known about cost-efficient strategies to control the disease. In an effort to address this gap in the literature, we model GLRD diffusion in a vineyard and evaluate bioeconomic outcomes under alternative disease control strategies.
We employ agent- based modeling ABM tools and contribute to bioeconomic literature on agricultural disease control in several ways. First, our model relaxes the assumption of agent homogeneity and allows instead agents to be heterogeneous in age and infection states, thus in their economic values. Second, we make the model inherently spatial-dynamic by combining the ABM with a cellular automaton system. Third, we incorporate realism when modeling the spread process by making the disease onset and its transmission stochastic.
That is, initial infections follow a random spatial distribution and stochastic agent interaction gives rise to Markov process- type disease diffusion. Finally, we formulate novel control strategies consisting of roguing and replacing infected grapevines based on their age and infection states. We evaluate these strategies and identify those that perform best at extending the expected vineyard half-life and at maximizing the vineyard expected net present values relative to the baseline of no control.
The model results underscore the ecological and economic tradeoffs implied by disease control strategies based on age and infection states. Friday, October 26, at 2: In this talk, we motivate the problem and present algorithms for probabilistic satisfiability PSATan NP-complete problem, focusing on the presence and absence of a phase transition phenomenon for each algorithm. Our study starts by defining a PSAT normal form, theo beijer exxonmobil fuels marketing which all algorithms are based.
Theoretical and practical limitations of each algorithm are discussed. Theo beijer exxonmobil fuels marketing algorithms are shown to present a phase transition behavior. We show that variations of these algorithms may lead to the partial occlusion of the phase transition phenomenon and discuss the reasons for this change of practical behavior. Friday, September 21, at 2: The high cost, limited capacity, and long recharge time of batteries pose a number of obstacles for the widespread adoption of electric vehicles.
Theo beijer exxonmobil fuels marketing systems that combine a standard battery with supercapacitors are currently one of the most promising theo beijer exxonmobil fuels marketing to increase battery lifespan and reduce operating costs. However, their performance crucially depends on how they are designed and operated. In this paper, we formalize the problem of optimizing real-time energy management of multi-battery systems as a stochastic planning problem, and we propose a novel solution based on a combination of optimization, machine learning and data-mining techniques.
We evaluate the performance of our intelligent energy management system on various large datasets of commuter trips crowdsourced in the United States. We show that our policy significantly outperforms the leading algorithms that were previously proposed as part of an open algorithmic challenge. We take the view that future advances in science, engineering, and medicine depend on the ability to comprehend the vast amounts of data being produced and acquired.
Visualization is a key enabling technology in this endeavor: Despite the promise that visualization can serve as an effective enabler of advances in other disciplines, the application of visualization technology is non-trivial.
The design of effective visualizations is a complex process that requires understanding of existing techniques and how they relate to human cognition. For a visualization to be insightful, it needs to be both effective and efficient. This requires a combination of design and science to reveal information that is otherwise obscured. In this talk, we will discuss recent work on the development of theo beijer exxonmobil fuels marketing visualization techniques and tools for a variety of needs.
He coauthored more than technical papers and eight U. Friday, April 20, at 2: Reintroducing Wildfire into Fire-Adapted Forests: Progress on Modeling the Let-Burn Decision. Forest Service theo beijer exxonmobil fuels marketing pursued a policy of aggressive fire suppression for the last century. In the fire-adapted forests of the western U. In this ICS project, we are bringing together existing models of forest vegetation development, fire behavior, fire suppression costs, and timber harvest to develop a policy rule to guide fire management planning in these forests.
So far, we have generated estimates of potential suppression cost savings arising from a let-burn decision. In this seminar, we report progress on developing a valuation function that combines suppression cost savings, timber harvest value at risk, and a reward function for restoration of pre-suppression-era forest conditions.
We will also report developments on how we plan to address the problem of optimal placement of fuel reducing treatments on the landscape. Friday, April 6, at 2: Decisions about the containment of harmful processes that spread across landscapes for example, wildfire and invasive species often must be made under uncertainty and as the system evolves in time.
Not all resources are available immediately and containment efforts may fail to prevent spread. The valuable probabilistic predictions produced by ecologists and foresters have been under-utilized because of the difficulty of optimizing when stochastic features and spatial connectedness interact.
I will introduce several simple models that generalize work in the CS theory literature and explain provably-good algorithmic results for some settings. These models capture qualitative tradeoffs with important implications for sustainable management. How should resources for wildfire containment be divided across preventive fuel removals and real-time fire suppression efforts, and how can these deployments be coordinated to maximum advantage?
If attempts to block invasive species spread are not perfectly reliable, but redundancy is costly, where should managers concentrate their resources? Friday, March 16, at 2: In environmental and natural resource planning, correlated management actions often need to be taken at a large number of locations over an extended period of time.
These problems present a variety of computational challenges for automated planning due to: In this talk I will present work from my Ph. I will also give an overview of how this approach fits in with other solution methods that are commonly used when trying to solve these planning problems.
In EPG the planning problem is modeled as a factored Markov decision process. A management policy is defined as a parameterized distribution over actions at an individual, generic location. The actual landscape policy is then the equilibrium distribution of a Markov chain built from these local policies.
In this framework the value model can contain local or global components as well as spatial constraints on actions taken at different locations. Also, the transition dynamics can be provided by existing simulators developed by domain experts.
EPG planning involves sampling from this equilibrium policy and iteratively adjusting the policy parameters using a policy gradient algorithm. Experiments using a BC Forest Service simulator demonstrate the algorithm's ability to devise policies for sustainable harvest planning of a forest with many locations. Mark Crowley completed his Ph. His research focuses on planning in large scale spatial domains such as forestry planning under the existence of uncertainty and spatial constraints.
A related research topic he looked at was how to compactly represent spatial policies as the equilibria of cyclic causal models and perform inference on those models. He also carried out research on shielding the unwanted effects of interventions used to model constraints in Bayesian networks. Theo beijer exxonmobil fuels marketing received his B.
