We study time-consistency of optimization problems, where we say that an optimization problem is time-consistent if its optimal solution, or the optimal policy for choosing actions, does not depend on when the optimization problem is solved. Time-consistency is a minimal requirement on an optimization problem for the decisions made based on its solution to be rational. We show that the return that we can gain by taking "optimal" actions selected by solving a time-inconsistent optimization problem can be surely dominated by that we could gain by taking "suboptimal" actions. We establish sufficient conditions on the objective function and on the constraints for an optimization problem to be time-consistent. We also show when the sufficient conditions are necessary. Our results are relevant in stochastic settings particularly when the objective function is a risk measure other than expectation or when there is a constraint on a risk measure.

Modeling trust in complex dynamic environments is an important yet challenging issue since an intelligent agent may strategically change its behavior to maximize its profits. In thispaper, we propose a context aware trust model to predict dynamic trust by using a Hidden Markov Model (HMM) to model an agent's interactions. Although HMMs have already been applied in the past to model an agent's dynamic behavior to greatly improve the traditional static probabilistic trust approaches, most HMM based trust models only focus on outcomes of the past interactions without considering interaction context, which we believe, reflects immensely on the dynamic behavior or intent of an agent. Interaction contextual information is comprehensively studied and integrated into the model to more precisely approximate an agent's dynamic behavior. Evaluation using real auction data and synthetic data demonstrates the efficacy of our approach in comparison with previous state-of-the-art trust mechanisms.

In extensive-form games with a large number of actions, careful abstraction of the action space is critically important to performance. In this paper we extend previous work on action abstraction using no-limit poker games as our test domains. We show that in such games it is no longer necessary to choose, a priori, one specific range of possible bet sizes. We introduce an algorithm that adjusts the range of bet sizes considered for each bet individually in an iterative fashion. This flexibility results in a substantially improved game value in no-limit Leduc poker. When applied to no-limit Texas Hold'em our algorithm produces an action abstraction that is about one third the size of a state of the art hand-crafted action abstraction, yet has a better overall game value.

We show that by modeling people as bounded finite automata, we can capture at a qualitative level the behavior observed in experiments. We consider a decision problem with incomplete information and a dynamically changing world, which can be viewed as an abstraction of many real-world settings. We provide a simple strategy for a finite automaton in this setting, and show that it does quite well, both through theoretical analysis and simulation. We show that, if the probability of nature changing state goes to 0 and the number of states in the automaton increases, then this strategy performs optimally (as well as if it were omniscient and knew when nature was making its state changes). Thus, although simple, the strategy is a sensible strategy for a resource-bounded agent to use. Moreover, at a qualitative level, the strategy does exactly what people have been observed to do in experiments.

We consider lifted importance sampling (LIS), a previously proposed approximate inference algorithm for statistical relational learning (SRL) models. LIS achieves substantial variance reduction over conventional importance sampling by using various lifting rules that take advantage of the symmetry in the relational representation. However, it suffers from two drawbacks. First, it does not take advantage of some important symmetries in the relational representation and may exhibit needlessly high variance on models having these symmetries. Second, it uses an uninformative proposal distribution which adversely affects its accuracy. We propose two improvements to LIS that address these limitations. First, we identify a new symmetry in SRL models and define a lifting rule for taking advantage of this symmetry. The lifting rule reduces the variance of LIS. Second, we propose a new, structured approach for constructing and dynamically updating the proposal distribution via adaptive sampling. We demonstrate experimentally that our new, improved LIS algorithm is substantially more accurate than the LIS algorithm.

Tractable subsets of first-order logic are a central topic in AI research. Several of these formalisms have been used as the basis for first-order probabilistic languages. However, these are intractable, losing the original motivation. Here we propose the first non-trivially tractable first-order probabilistic language. It is a subset of Markov logic, and uses probabilistic class and part hierarchies to control complexity. We call it TML (Tractable Markov Logic). We show that TML knowledge bases allow for efficient inference even when the corresponding graphical models have very high treewidth. We also show how probabilistic inheritance, default reasoning, and other inference patterns can be carried out in TML. TML opens up the prospect of efficient large-scale first-order probabilistic inference.

The paper focuses on finding the m best solutions to combinatorial optimization problems using Best-First or Branchand- Bound search. Specifically, we present m-A*, extending the well-known A* to the m-best task, and prove that all its desirable properties, including soundness, completeness and optimal efficiency, are maintained. Since Best-First algorithms have memory problems, we also extend the memoryefficient Depth-First Branch-and-Bound to the m-best task. We extend both algorithms to optimization tasks over graphical models (e.g., Weighted CSP and MPE in Bayesian networks), provide complexity analysis and an empirical evaluation. Our experiments with 5 variants of Best-First and Branch-and-Bound confirm that Best-First is largely superior when memory is available, but Branch-and-Bound is more robust, while both styles of search benefit greatly when the heuristic evaluation function has increased accuracy.

This paper concerns learning and prediction with probabilistic models where the domain sizes of latent variables have no a priori upper-bound. Current approaches represent prior distributions over latent variables by stochastic processes such as the Dirichlet process, and rely on Monte Carlo sampling to estimate the model from data. We propose an alternative approach that searches over the domain size of latent variables, and allows arbitrary priors over the their domain sizes. We prove error bounds for expected probabilities, where the error bounds diminish with increasing search scope. The search algorithm can be truncated at any time . We empirically demonstrate the approach for topic modelling of text documents.

We study the marginal-MAP problem on graphical models, and present a novel approximation method based on direct approximation of the sum operation. A primary difficulty of marginal-MAP problems lies in the non-commutativity of the sum and max operations, so that even in highly structured models, marginalization may produce a densely connected graph over the variables to be maximized, resulting in an intractable potential function with exponential size. We propose a chain decomposition approach for summing over the marginalized variables, in which we produce a structured approximation to the MAP component of the problem consisting of only pairwise potentials. We show that this approach is equivalent to the maximization of a specific variational free energy, and it provides an upper bound of the optimal probability. Finally, experimental results demonstrate that our method performs favorably compared to previous methods.

The presence of non-symmetric evidence has been a barrier for the application of lifted inference since the evidence destroys the symmetry of the first-order probabilistic model. In the extreme case, if distinct soft evidence is obtained about each individual object in the domain then, often, all current exact lifted inference methods reduce to traditional inference at the ground level. However, it is of interest to ask whether the symmetry of the model itself before evidence is obtained can be exploited. We present new results showing that this is, in fact, possible. In particular, we show that both exact maximum a posteriori (MAP) and marginal inference can be lifted for the case of distinct soft evidence on a unary Markov Logic predicate. Our methods result in efficient procedures for MAP and marginal inference for a class of graphical models previously thought to be intractable.