Inspired by “GoogleTM Sets” and Bayesian sets, we consider the problem of retrieving complex objects and relations among them, i.e., ground atoms from a logical concept, given a query consisting of a few atoms from that concept. We formulate this as a within-network relational learning problem using few labels only and describe an algorithm that ranks atoms using a score based on random walks with restart (RWR): the probability that a random surfer hits an atom starting from the query atoms. Specifically, we compute an initial ranking using personalized PageRank. Then, we find paths of atoms that are connected via their arguments, variablize the ground atoms in each path, in order to create features for the query. These features are used to re-personalize the original RWR and to finally compute the set completion, based on Label Propagation. Moreover, we exploit that RWR techniques can naturally be lifted and show that lifted inference for label propagation is possible. We evaluate our algorithm on a realworld relational dataset by finding completions of sets of objects describing the Roman city of Pompeii. We compare to Bayesian sets and show that our approach gives very reasonable set completions.

Since learning in Boltzmann machines is typically quite slow, there is a need to restrict connections within hidden layers. However, theresulting states of hidden units exhibit statistical dependencies. Based on this observation, we propose using l1/l2 regularization upon the activation probabilities of hidden units in restricted Boltzmann machines to capture the local dependencies among hidden units. This regularization not only encourages hidden units of many groups to be inactive given observed data but also makes hidden units within a group compete with each other for modeling observed data. Thus, the l1/l2 regularization on RBMs yields sparsity at both the group and the hidden unit levels. We call RBMs trained with the regularizer sparse group RBMs (SGRBMs). The proposed SGRBMs are appliedto model patches of natural images, handwritten digits and OCR English letters. Then to emphasize that SGRBMs can learn more discriminative features we applied SGRBMs to pretrain deep networks for classification tasks. Furthermore, we illustrate the regularizer can also be applied to deep Boltzmann machines, which lead to sparse group deep Boltzmann machines. When adapted to the MNIST data set, a two-layer sparse group Boltzmann machine achieves an error rate of 0.84%, which is, to our knowledge, the best published result on the permutation-invariant version of the MNIST task.

Recently, a number of researchers have proposed spectral algorithms for learning models of dynamical systems — for example, Hidden Markov Models (HMMs), Partially Observable Markov Decision Processes (POMDPs), and Transformed Predictive State Representations (TPSRs). These algorithms are attractive since they are statistically consistent and not subject to local optima. However, they are batch methods: they need to store their entire training data set in memory at once and operate on it as a large matrix, and so they cannot scale to extremely large data sets (either many examples or many features per example). In turn, this restriction limits their ability to learn accurate models of complex systems. To overcome these limitations, we propose a new online spectral algorithm, which uses tricks such as incremental Singular Value Decomposition (SVD) and random projections to scale to much larger data sets and more complex systems than previous methods. We demonstrate the new method on an inertial measurement prediction task and a high-bandwidth video mapping task and we illustrate desirable behaviors such as "closing the loop," where the latent state representation changes suddenly as the learner recognizes that it has returned to a previously known place.

This paper addresses one of the key components of the mining process: the geological prediction of natural resources from spatially distributed measurements. We present a novel approach combining undirected graphical models with ensemble classifiers to provide 3D geological models from multiple sensors installed in an autonomous drill rig. Drill sensor measurements used for drilling automation, known as measurement-while-drilling (MWD) data, have the potential to provide an estimate of the geological properties of the rocks being drilled. The proposed method maps MWD parameters to rock types while considering spatial relationships, i.e., associating measurements obtained from neighboring regions. We use a conditional random field with local information provided by boosted decision trees to jointly reason about the rock categories of neighboring measurements. To validate the approach, MWD data was collected from a drill rig operating at an iron ore mine. Graphical models of the 3D structure present in real data sets possess a high number of nodes, edges and cycles, making them intractable for exact inference. We provide a comparison of three approximate inference methods to calculate the most probable distribution of class labels. The empirical results demonstrate the benefits of spatial modeling through graphical models to improve classification performance.

Markov Decision Processes arise as a natural model for many renewable resources allocation problems. In many such problems, high stakes decisions with potentially catastrophic outcomes (such as the collapse of an entire ecosystem) need to be taken by carefully balancing social, economic, and ecologic goals. We introduce a broad class of such MDP models with a risk averse attitude of the decision maker, in order to obtain policies that are more balanced with respect to the welfare of future generations. We prove that they admit a closed form solution that can be efficiently computed. We show an application of the proposed framework to the Pacific Halibut marine fishery, obtaining new and more cautious policies. Our results strengthen findings of related policies from the literature by providing new evidence that a policy based on periodic closures of the fishery should be employed, in place of the one traditionally used that harvests a constant proportion of the stock every year.

Log-linear description logics are a family of probabilistic logics integrating various concepts and methods from the areas of knowledge representation and reasoning and statistical relational AI. We define the syntax and semantics of log-linear description logics, describe a convenient representation as sets of first-order formulas, and discuss computational and algorithmic aspects of probabilistic queries in the language. The paper concludes with an experimental evaluation of an implementation of a log-linear DL reasoner.

Topic models have been used successfully for a variety of problems, often in the form of application-specific extensions of the basic Latent Dirichlet Allocation (LDA) model. Because deriving these new models in order to encode domain knowledge can be difficult and time-consuming, we propose the Fold·all model, which allows the user to specify general domain knowledge in First-Order Logic (FOL). However, combining topic modeling with FOL can result in inference problems beyond the capabilities of existing techniques. We have therefore developed a scalable inference technique using stochastic gradient descent which may also be useful to the Markov Logic Network (MLN) research community. Experiments demonstrate the expresive power of Fold·all, as well as the scalability of our proposed inference method.

PODTGolog [Rens, 2010] is a Golog dialect attempting Broadly speaking, my research concerns combining to deal with partially observable MDP (POMDP) logic of action and POMDP theory in a coherent, environments. PODTGolog has not been given a mathematical theoretically sound language for agent programming.

The scalability of algorithms for solving Markov Decision Processes (MDPs) has been a limiting factor for MDPs as a modeling tool. This dissertation develops theoretical and empirical techniques for solving larger MDPs than was possible before, and aims to demonstrate the achieved progress by applying these new algorithms to a real-world problem.

Human Robot Interaction (HRI) is an active field of integrating and embedding different techniques in artificial intelligence. This paper describes my research topic on: Control of Robotic Systems for Safe Interaction with Human Operators. It consists of online motion generation for robotic manipulators interacting with dynamic obstacles and humans using a moving horizon scheme, modeling and long term prediction of human motion using probabilistic models and reachability analysis, and development of an HRI demonstration platform.