Recent research leverages results from the continuous-armed bandit literature to create a reinforcement-learning algorithm for continuous state and action spaces. Initially proposed in a theoretical setting, we provide the first examination of the empirical properties of the algorithm. Through experimentation, we demonstrate the effectiveness of this planning method when coupled with exploration and model learning and show that, in addition to its formal guarantees, the approach is very competitive with other continuous-action reinforcement learners.

The problem of automated planning is known to be intractable in general. Moreover, it has been proven that in some cases finding an optimal solution is much harder than finding any solution. Existing techniques have to compromise between speed of the planning process and quality of solutions. For example, techniques based on greedy search often are able to obtain solutions quickly, but the quality of the solutions is usually low. Similarly, adding macro-operators to planning domains often enables planning speed-up, but solution sequences are typically longer. In this paper, we propose a method for optimizing plans with respect to their length, by post-planning analysis. The method is based on analyzing action dependencies and independencies by which we are able to identify redundant actions or non-optimal sub-plans. To evaluate the process we provide preliminary empirical evidence using benchmark domains.

Human users who execute an automatically generated plan want to understand the rationale behind it. Knowledge-rich plans are particularly suitable for this purpose, because they provide the means to give reason for causal, temporal, and hierarchical relationships between actions. Based on this information, focused arguments can be generated that constitute explanations on an appropriate level of abstraction. In this paper, we present a formal approach to plan explanation. Information about plans is represented as first-order logic formulae and explanations are constructed as proofs in the resulting axiomatic system. With that, plan explanations are provably correct w.r.t. the planning system that produced the plan. A prototype plan explanation system implements our approach and first experiments give evidence that finding plan explanations is feasible in real-time.

This paper proposes a new algorithm for multiple sparse regression in high dimensions, where the task is to estimate the support and values of several (typically related) sparse vectors from a few noisy linear measurements. Our algorithm is a "forward-backward" greedy procedure that -- uniquely -- operates on two distinct classes of objects. In particular, we organize our target sparse vectors as a matrix; our algorithm involves iterative addition and removal of both (a) individual elements, and (b) entire rows (corresponding to shared features), of the matrix. Analytically, we establish that our algorithm manages to recover the supports (exactly) and values (approximately) of the sparse vectors, under assumptions similar to existing approaches based on convex optimization. However, our algorithm has a much smaller computational complexity. Perhaps most interestingly, it is seen empirically to require visibly fewer samples. Ours represents the first attempt to extend greedy algorithms to the class of models that can only/best be represented by a combination of component structural assumptions (sparse and group-sparse, in our case).

We present a probabilistic viewpoint to multiple kernel learning unifying well-known regularised risk approaches and recent advances in approximate Bayesian inference relaxations. The framework proposes a general objective function suitable for regression, robust regression and classification that is lower bound of the marginal likelihood and contains many regularised risk approaches as special cases. Furthermore, we derive an efficient and provably convergent optimisation algorithm.

Agricultural research has been profited by technical advances such as automation, data mining. Today, data mining is used in a vast areas and many off-the-shelf data mining system products and domain specific data mining application soft wares are available, but data mining in agricultural soil datasets is a relatively a young research field. The large amounts of data that are nowadays virtually harvested along with the crops have to be analyzed and should be used to their full extent. This research aims at analysis of soil dataset using data mining techniques. It focuses on classification of soil using various algorithms available. Another important purpose is to predict untested attributes using regression technique, and implementation of automated soil sample classification.

We study the generalization performance of online learning algorithms trained on samples coming from a dependent source of data. We show that the generalization error of any stable online algorithm concentrates around its regret--an easily computable statistic of the online performance of the algorithm--when the underlying ergodic process is $\beta$- or $\phi$-mixing. We show high probability error bounds assuming the loss function is convex, and we also establish sharp convergence rates and deviation bounds for strongly convex losses and several linear prediction problems such as linear and logistic regression, least-squares SVM, and boosting on dependent data. In addition, our results have straightforward applications to stochastic optimization with dependent data, and our analysis requires only martingale convergence arguments; we need not rely on more powerful statistical tools such as empirical process theory.

We find lower and upper bounds for the risk of estimating a manifold in Hausdorff distance under several models. We also show that there are close connections between manifold estimation and the problem of deconvolving a singular measure.

In a regression setup with deterministic design, we study the pure aggregation problem and introduce a natural extension from the Gaussian distribution to distributions in the exponential family. While this extension bears strong connections with generalized linear models, it does not require identifiability of the parameter or even that the model on the systematic component is true. It is shown that this problem can be solved by constrained and/or penalized likelihood maximization and we derive sharp oracle inequalities that hold both in expectation and with high probability. Finally all the bounds are proved to be optimal in a minimax sense.

Term weighting schemes often dominate the performance of many classifiers, such as kNN, centroid-based classifier and SVMs. The widely used term weighting scheme in text categorization, i.e., tf.idf, is originated from information retrieval (IR) field. The intuition behind idf for text categorization seems less reasonable than IR. In this paper, we introduce inverse category frequency (icf) into term weighting scheme and propose two novel approaches, i.e., tf.icf and icf-based supervised term weighting schemes. The tf.icf adopts icf to substitute idf factor and favors terms occurring in fewer categories, rather than fewer documents. And the icf-based approach combines icf and relevance frequency (rf) to weight terms in a supervised way. Our cross-classifier and cross-corpus experiments have shown that our proposed approaches are superior or comparable to six supervised term weighting schemes and three traditional schemes in terms of macro-F1 and micro-F1.