Utility elicitation is an important component of many applications, such as decision support systems and recommender systems. Such systems query users about their preferences and offer recommendations based on the system's belief about the user's utility function. We analyze the connection between the problem of generating optimal recommendation sets and the problem of generating optimal choice queries, considering both Bayesian and regret-based elicitation. Our results show that, somewhat surprisingly, under very general circumstances, the optimal recommendation set coincides with the optimal query.

The focus of my research is an agent-based system for optimizing spatial arrangements of plants on a landscape to maximize their growth and minimize their water use. The optimization criteria include a natural phenomenon known as facilitation, which is observed in water-scarce environments when larger shrubs serve as benefactors to smaller annuals by generating conditions that protect them from harsh afternoon sun. In my modeling and optimization system each plant is an agent with growth requirements. A plant agent's fitness at a given location is defined by a fitness function that includes those growth requirements and a penalty term designed to force facilitation. The landscape design is formulated as a combinatorial optimization problem with a discrete set of locations for each plant on a grid, a fixed number of plants, and a fitness function that defines the performance of a plant at a location. To evaluate the effectiveness of this approach, I applied a variety of search strategies, including simulated annealing and a new agent-based approach that mimics how plant communities evolve over time, to different collections of simulated plant types and landscapes and compared the fitness scores and spatial arrangments in the solutions. The fitness scores from the search strategies were comparable. The search strategies produced different spatial distributions of the larger plants, and all designs exhibited facilitation and lower water use.

Combining Answer Set Programming (ASP) and Constraint Logic Programming (CLP) can create a more powerful language for knowledge representation and reasoning. The language AC(C) is designed to integrate ASP and CLP. Compared with existing integration of ASP and CSP, AC(C) allows representing user-defined constraints. Such integration provides great power for applications requiring logical reasoning involving constraints, e.g., temporal planning. In AC(C), user-defined and primitive constraints can be solved by a CLP inference engine while the logical reasoning over those constraints and regular logic literals is solved by an ASP inference engine (i.e., solver). My PhD work includes improving the language AC(C), implementing its faster inference engine and investigating how effective the new system can be used to solve a challenging application, temporal planning.

Search algorithms often suffer from exploring areas which eventually are not part of the shortest path from the start to a goal. Usually it is the purpose of the heuristic function to guide the search algorithm such that it will ignore as much as possible of these areas. We consider other, non-heuristic methods that can be used to prune the search space to make search even faster. We present two algorithms: one for search in graphs that fit in memory, and in which we will need to perform many searches, and another, which improves the search time of planning problems that contain symmetries.

Multi-task learning has received increasing attention in the past decade. Many supervised multi-task learning methods have been proposed, while unsupervised multi-task learning is still a rarely studied problem. In this paper, we propose to learn a kernel for multi-task clustering. Our goal is to learn a Reproducing Kernel Hilbert Space, in which the geometric structure of the data in each task is preserved, while the data distributions of any two tasks are as close as possible. This is formulated as a unified kernel learning framework, under which we study two types of kernel learning: nonparametric kernel learning and spectral kernel design. Both types of kernel learning can be solved by linear programming. Experiments on several cross-domain text data sets demonstrate that kernel k-means on the learned kernel can achieve better clustering results than traditional single-task clustering methods. It also outperforms the newly proposed multi-task clustering method.

My work aims to build a system to automatically predict the writing quality in scientific articles from two genres—academic publications and science journalism. Our goal is to employ these predictions for article recommendation systems and to provide feedback during writing.

Cooperative path-finding can be abstracted as computing non-colliding paths for multiple agents between their start and goal locations on a graph. This work proposes a fast algorithm that can provide completeness guarantees for a general class of problems without any assumptions about the graph's topology. Specifically, the approach can address any solvable instance where there are at most n-2 agents in a graph of size n. The algorithm employs two primitives: a "push" operation where agents move towards their goals up to the point that no progress can be made, and a "swap" operation that allows two agents to swap positions without altering the configuration of other agents. Simulated experiments are provided on hard instances of cooperative path-finding, including comparisons against alternative methods. The results are favorable for the proposed algorithm and show that the technique scales to problems that require high levels of coordination, involving hundreds of agents.

We consider settings in which a semi-autonomous agent has uncertain knowledge about its environment, but can ask what action the human operator would prefer taking in the current or in a potential future state. Asking queries can improve behavior, but if queries come at a cost (e.g., due to limited operator attention), the value of each query should be maximized. We compare two strategies for selecting action queries: 1) based on myopically maximizing expected gain in long-term value, and 2) based on myopically minimizing uncertainty in the agent's policy representation. We show empirically that the first strategy tends to select more valuable queries, and that a hybrid method can outperform either method alone in settings with limited computation.

In environmental and natural resource planning domains actions are taken at a large number of locations over multiple time periods. These problems have enormous state and action spaces, spatial correlation between actions, uncertainty and complex utility models. We present an approach for modeling these planning problems as factored Markov decision processes. The reward model can contain local and global components as well as spatial constraints between locations. The transition dynamics can be provided by existing simulators developed by domain experts. We propose a landscape policy defined as the equilibrium distribution of a Markov chain built from many locally-parameterized policies. This policy is optimized using a policy gradient algorithm. Experiments using a forestry simulator demonstrate the algorithm's ability to devise policies for sustainable harvest planning of a forest.

Recent advances in click modeling have established it as an attractive approach to interpret search click data. These advances characterize users' search behavior either in advertisement blocks, or within an organic search block through probabilistic models. Yet, when searching for information on a search result page, one is often interacting with the search engine via an entire page instead of a single block. Consequently, previous works that exclusively modeled user behavior in a single block may sacrifice much useful user behavior information embedded in other blocks. To solve this problem, in this paper, we put forward a novel Whole Page Click (WPC) Model to characterize user behavior in multiple blocks. Specifically, WPC uses a Markov chain to learn the user transition probabilities among different blocks in the whole page. To compare our model with the best alternatives in the Web-Search literature, we run a large-scale experiment on a real dataset and demonstrate the advantage of the WPC model in terms of both the whole page and each block in the page. Especially, we find that WPC can achieve significant gain in interpreting the advertisement data, despite of the sparsity of the advertisement click data.