GENERATION by SHU ZHANG (Under the Direction of Khaled Rasheed) ABSTRACT In this thesis, two music accompaniment systems are presented. Evac (the evolutionary accompanist) is a system that engages in musical improvisation with the user. It uses a novel, implicitly interactive, genetic algorithm (GA), which allows the user’s actions to influence Evac’s musical performance without the need for ex...

Maximization of submodular functions has wide applications in machine learning and artificial intelligence. Adaptive submodular maximization has been traditionally studied under the assumption that the model of the world, the expected gain of choosing an item given previously selected items and their states, is known. In this paper, we study the setting where the expected gain is initially unkn...

This paper investigates the problem of policy search based on the only expert’s preferences. Whereas reinforcement learning classically relies on a reward function, or exploits the expert’s demonstrations, preference-based policy learning (PPL) iteratively builds and optimizes a policy return estimate as follows: The learning agent demonstrates a few policies, is informed of the expert’s prefer...

It is well-known from psycholinguistic literature that the human language processing system exhibits preferences when sentence constituents are ambiguous with respect to their grammatical function. Generally, many theories assume that an interpretation towards the subject is preferred in such cases. Later disambiguations which contradict such a preference induce enhanced processing difficulty (...

Objective: Methamphetamine (METH) increases dopamine, norepinephrine and serotonin concentrations in the synaptic cleft, and induces hyperactivity. The current management of acute METH poisoning relies on supportive care and no specific antidote is available for treatment. The main objective of this review was to present the evidence for effectiveness of the herbal medicine in ...

A four-attribute health state classification system designed to uniquely categorize the health status of all individuals two years of age and over is presented. A social preference function defined over the health state classification system is required. Standard multi-attribute utility theory is investigated for the task, problems are identified and modifications to the standard method are pro...

Given a finite set of generalized lotteries, that is random quantities equipped with a belief function, and a partial preference relation on them, a necessary and sufficient condition (Choquet rationality) has been provided for its representation as a Choquet expected utility of a strictly increasing utility function. Here we prove that this condition assures the extension of the preference rel...

Many real-world applications are characterized by a number of conflicting performance measures. As optimizing in a multi-objective setting leads to a set of non-dominated solutions, a preference function is required for selecting the solution with the appropriate trade-off between the objectives. This preference function is often unknown, especially when it comes from an expert human user. Howe...

Distributed Constraint Optimization Problems (DCOPs) offer a powerful approach for the description and resolution of cooperative multi-agent problems. In such model a group of agents coordinate their actions to optimize a global objective function, taking into account their preferences and constraints. A core limitation of this model is the assumption that all agents’ preferences are specified ...

There are two key components to constructing summary measures of population health quality: a health-status classification system and a valuation function. The former was discussed in part 7. My remarks on that session are also relevant here (see chapter 7.2). The latter will be the focus of my brief remarks here. I will concentrate on three major topics: a conceptual foundation that supports a...