There are various algorithms for finding a Bayesian networkstructure (BNS) that is optimal with respect to a given scoring function. No single algorithm dominates the others in speed, and, given a problem instance, it is a priori unclear which algorithm will perform best and how fast it will solve the problem. Estimating the runtimes directly is extremely difficult as they are complicated functions of the instance. The main contribution of this paper is characterization of the empirical hardness of an instance for a given algorithm based on a novel collection of non-trivial, yet efficiently computable features. Our empirical results, based on the largest evaluation of state-of-the-art BNS learning algorithms to date, demonstrate that we can predict the runtimes to a reasonable degree of accuracy, and effectively select algorithms that perform well on a particular instance. Moreover, we also show how the results can be utilized in building a portfolio algorithm that combines several individual algorithms in an almost optimal manner.

Selection bias is caused by preferential exclusion of units from the samples and represents a major obstacle to valid causal and statistical inferences; it cannot be removed by randomized experiments and can rarely be detected in either experimental or observational studies. In this paper, we provide complete graphical and algorithmic conditions for recovering conditional probabilities from selection biased data. We also provide graphical conditions for recoverability when unbiased data is available over a subset of the variables. Finally, we provide a graphical condition that generalizes the backdoor criterion and serves to recover causal effects when the data is collected under preferential selection.

Transfer learning considers related but distinct tasks defined on heterogenous domains and tries to transfer knowledge between these tasks to improve generalization performance. It is particularly useful when we do not have sufficient amount of labeled training data in some tasks, which may be very costly, laborious, or even infeasible to obtain. Instead, learning the tasks jointly enables us to effectively increase the amount of labeled training data. In this paper, we formulate a kernelized Bayesian transfer learning framework that is a principled combination of kernel-based dimensionality reduction models with task-specific projection matrices to find a shared subspace and a coupled classification model for all of the tasks in this subspace. Our two main contributions are: (i) two novel probabilistic models for binary and multiclass classification, and (ii) very efficient variational approximation procedures for these models. We illustrate the generalization performance of our algorithms on two different applications. In computer vision experiments, our method outperforms the state-of-the-art algorithms on nine out of 12 benchmark supervised domain adaptation experiments defined on two object recognition data sets. In cancer biology experiments, we use our algorithm to predict mutation status of important cancer genes from gene expression profiles using two distinct cancer populations, namely, patient-derived primary tumor data and in-vitro-derived cancer cell line data. We show that we can increase our generalization performance on primary tumors using cell lines as an auxiliary data source.

This paper presents an extension of the cascading Indian buffet process (CIBP) intended to learning arbitrary directed acyclic graph structures as opposed to the CIBP, which is limited to purely layered structures. The extended cascading Indian buffet process (eCIBP) essentially consists in adding an extra sampling step to the CIBP to generate connections between non-consecutive layers. In the context of graphical model structure learning, the proposed approach allows learning structures having an unbounded number of hidden random variables and automatically selecting the model complexity. We evaluated the extended process on multivariate density estimation and structure identification tasks by measuring the structure complexity and predictive performance. The results suggest the extension leads to extracting simpler graphs without scarifying predictive precision.

Qualitative Possibilistic Mixed-Observable MDPs (pi-MOMDPs), generalizing pi-MDPs and pi-POMDPs, are well-suited models to planning under uncertainty with mixed-observability when transition, observation and reward functions are not precisely known and can be qualitatively described. Functions defining the model as well as intermediate calculations are valued in a finite possibilistic scale L, which induces a finite belief state space under partial observability contrary to its probabilistic counterpart. In this paper, we propose the first study of factored pi-MOMDP models in order to solve large structured planning problems under qualitative uncertainty, or considered as qualitative approximations of probabilistic problems. Building upon the SPUDD algorithm for solving factored (probabilistic) MDPs, we conceived a symbolic algorithm named PPUDD for solving factored pi-MOMDPs. Whereas SPUDD's decision diagrams' leaves may be as large as the state space since their values are real numbers aggregated through additions and multiplications, PPUDD's ones always remain in the finite scale L via min and max operations only. Our experiments show that PPUDD's computation time is much lower than SPUDD, Symbolic-HSVI and APPL for possibilistic and probabilistic versions of the same benchmarks under either total or mixed observability, while still providing high-quality policies.

In the field of NLP, most of the existing domain adaptation studies belong to the feature-based adaptation, while the research of instance-based adaptation is very scarce. In this work, we propose a new instance-based adaptation model, called in-target-domain logistic approximation (ILA). In ILA, we adapt the source-domain data to the target domain by a logistic approximation. The normalized in-target-domain probability is assigned as an instance weight to each of the source-domain training data. An instance-weighted classification model is trained finally for the cross-domain classification problem. Compared to the previous techniques, ILA conducts instance adaptation in a dimensionality-reduced linear feature space to ensure efficiency in high-dimensional NLP tasks. The instance weights in ILA are learnt by leveraging the criteria of both maximum likelihood and minimum statistical distance. The empirical results on two NLP tasks including text categorization and sentiment classification show that our ILA model beats the state-of-the-art instance adaptation methods significantly, in cross-domain classification accuracy, parameter stability and computational efficiency.

In this paper we develop an algorithm to find the k-best equivalence classes of Bayesian networks. Our algorithm is capable of finding much more best DAGs than the previous algorithm that directly finds the k-best DAGs (Tian, He and Ram 2010). We demonstrate our algorithm in the task of Bayesian model averaging. Empirical results show that our algorithm significantly outperforms the k-best DAG algorithm in both time and space to achieve the same quality of approximation. Our algorithm goes beyond the maximum-a-posteriori (MAP) model by listing the most likely network structures and their relative likelihood and therefore has important applications in causal structure discovery.

Collaborative filtering algorithms generally rely on the assumption that user preference patterns remain stationary. However, real-world relational data are seldom stationary. User preference patterns may change over time, giving rise to the requirement of designing collaborative filtering systems capable of detecting and adapting to preference pattern shifts. Motivated by this observation, in this paper we propose a dynamic Bayesian probabilistic matrix factorization model, designed for modeling time-varying distributions. Formulation of our model is based on imposition of a dynamic hierarchical Dirichlet process (dHDP) prior over the space of probabilistic matrix factorization models to capture the time-evolving statistical properties of modeled sequential relational datasets. We develop a simple Markov Chain Monte Carlo sampler to perform inference. We present experimental results to demonstrate the superiority of our temporal model.

Indeed, to the best of our knowledge, we lack forgotten over time (Wu and Humberman 2007). For example, a probabilistic framework to model and predict the popularity videos on YouTube or stories on Digg gain their popularity dynamics of individual items. The reason behind this is by striving for views or votes (Szabo and Huberman partly illustrated in Figure 1, suggesting that the dynamical 2010); papers increase their visibility by competing for citations processes governing individual items appear too noisy to be from new papers (Ren et al. 2010; Wang, Song, and amenable to quantification. Barabási 2013); tweets or Hashtags in Twitter become more In this paper, we model the stochastic popularity dynamics popular as being retweeted (Hong, Dan, and Davison 2011) using reinforced Poisson processes, capturing simultaneously and so do webpages as being attached by incoming hyperlinks three key ingredients: fitness of an item, characterizing (Ratkiewicz et al. 2010). An ability to predict the popularity its inherent competitiveness against other items; a general of individual items within a dynamically evolving system temporal relaxation function, corresponding to the aging not only probes our understanding of complex systems, in the ability to attract new attentions; and a reinforcement but also has important implications in a wide range of domains, mechanism, documenting the well-known "rich-get-richer" from marketing and traffic control to policy making phenomenon. The benefit of the proposed model is threefold: and risk management. Despite recent advances of empirical (1) It models the arrival process of individual attentions methods, we lack a general modeling framework to predict directly in contrast to relying on aggregated popularity the popularity of individual items within a complex evolving time series; (2) As a generative probabilistic model, it can be system.

Recent works have explored the use of brain signals to directly control virtual and robotic agents in sequential tasks. So far in such brain-computer interfaces (BCI), an explicit calibration phase was required to build a decoder that translates raw electroencephalography (EEG) signals from the brain of each user into meaningful instructions. This paper proposes a method that removes the calibration phase, and allows a user to control an agent to solve a sequential task. The proposed method assumes a distribution of possible tasks, and infers the interpretation of EEG signals and the task by selecting the hypothesis which best explains the history of interaction. We introduce a measure of uncertainty on the task and on the EEG signal interpretation to act as an exploratory bonus for a planning strategy. This speeds up learning by guiding the system to regions that better disambiguate among task hypotheses. We report experiments where four users use BCI to control an agent on a virtual world to reach a target without any previous calibration process.