There has been a clear distinction between induction or training time and diagnosis time active information acquisition. While active learning during induction focuses on acquiring data that promises to provide the best classification model, the goal at diagnosis time focuses completely on next features to observe about the test case at hand in order to make better predictions about the case. We introduce a model and inferential methods that breaks this distinction. The methods can be used to extend case libraries under a budget but, more fundamentally, provide a framework for guiding agents to collect data under scarce resources, focused by diagnostic challenges. This extension to active learning leads to a new class of policies for real-time diagnosis, where recommended information-gathering sequences include actions that simultaneously seek new data for the case at hand and for cases in the training set.

Research in animal learning and behavioral neuroscience has distinguished between two forms of action control: a habit-based form, which relies on stored action values, and a goal-directed form, which forecasts and compares action outcomes based on a model of the environment. While habit-based control has been the subject of extensive computational research, the computational principles underlying goal-directed control in animals have so far received less attention. In the present paper, we advance a computational framework for goal-directed control in animals and humans. We take three empirically motivated points as founding premises: (1) Neurons in dorsolateral prefrontal cortex represent action policies, (2) Neurons in orbitofrontal cortex represent rewards, and (3) Neural computation, across domains, can be appropriately understood as performing structured probabilistic inference. On a purely computational level, the resulting account relates closely to previous work using Bayesian inference to solve Markov decision problems, but extends this work by introducing a new algorithm, which provably converges on optimal plans. On a cognitive and neuroscientific level, the theory provides a unifying framework for several different forms of goal-directed action selection, placing emphasis on a novel form, within which orbitofrontal reward representations directly drive policy selection.

We propose Dirichlet-Bernoulli Alignment (DBA), a generative model for corpora in which each pattern (e.g., a document) contains a set of instances (e.g., paragraphs in the document) and belongs to multiple classes. By casting predefined classes as latent Dirichlet variables (i.e., instance level labels), and modeling the multi-label of each pattern as Bernoulli variables conditioned on the weighted empirical average of topic assignments, DBA automatically aligns the latent topics discovered from data to human-defined classes. DBA is useful for both pattern classification and instance disambiguation, which are tested on text classification and named entity disambiguation for web search queries respectively.

Classical Game Theoretic approaches that make strong rationality assumptions have difficulty modeling observed behaviour in Economic games of human subjects. We investigate the role of finite levels of iterated reasoning and non-selfish utility functions in a Partially Observable Markov Decision Process model that incorporates Game Theoretic notions of interactivity. Our generative model captures a broad class of characteristic behaviours in a multi-round Investment game. We invert the generative process for a recognition model that is used to classify 200 subjects playing an Investor-Trustee game against randomly matched opponents.

We describe a novel class of distributions, called Mondrian processes, which can be interpreted as probability distributions over kd-tree data structures. Mondrian processesare multidimensional generalizations of Poisson processes and this connection allows us to construct multidimensional generalizations of the stickbreaking processdescribed by Sethuraman (1994), recovering the Dirichlet process in one dimension. After introducing the Aldous-Hoover representation for jointly and separately exchangeable arrays, we show how the process can be used as a nonparametric prior distribution in Bayesian models of relational data.

Conditional random fields (CRF) are quite successful on sequence labeling tasks such as natural language processing and biological sequence analysis. CRF models use linear potential functions to represent the relationship between input features and outputs. However, in many real-world applications such as protein structure prediction and handwriting recognition, the relationship between input features and outputs is highly complex and nonlinear, which cannot be accurately modeled by a linear function. To model the nonlinear relationship between input features and outputs we propose Conditional Neural Fields (CNF), a new conditional probabilistic graphical model for sequence labeling. Our CNF model extends CRF by adding one (or possibly several) middle layer between input features and outputs. The middle layer consists of a number of hidden parameterized gates, each acting as a local neural network node or feature extractor to capture the nonlinear relationship between input features and outputs. Therefore, conceptually this CNF model is much more expressive than the linear CRF model. To better control the complexity of the CNF model, we also present a hyperparameter optimization procedure within the evidence framework. Experiments on two widely-used benchmarks indicate that this CNF model performs significantly better than a number of popular methods. In particular, our CNF model is the best among about ten machine learning methods for protein secondary tructure prediction and also among a few of the best methods for handwriting recognition.

We introduce the first temporal-difference learning algorithm that is stable with linear function approximation and off-policy training, for any finite Markov decision process, target policy, and exciting behavior policy, and whose complexity scales linearly in the number of parameters. We consider an i.i.d.\ policy-evaluation setting in which the data need not come from on-policy experience. The gradient temporal-difference (GTD) algorithm estimates the expected update vector of the TD(0) algorithm and performs stochastic gradient descent on its L_2 norm. Our analysis proves that its expected update is in the direction of the gradient, assuring convergence under the usual stochastic approximation conditions to the same least-squares solution as found by the LSTD, but without its quadratic computational complexity. GTD is online and incremental, and does not involve multiplying by products of likelihood ratios as in importance-sampling methods.

We define a metric for measuring behavior similarity between states in a Markov decision process (MDP), which takes action similarity into account. We show that the kernel of our metric corresponds exactly to the classes of states defined by MDP homomorphisms (Ravindran & Barto, 2003). We prove that the difference inthe optimal value function of different states can be upper-bounded by the value of this metric, and that the bound is tighter than previous bounds provided bybisimulation metrics (Ferns et al. 2004, 2005). Our results hold both for discrete and for continuous actions. We provide an algorithm for constructing approximate homomorphisms, by using this metric to identify states that can be grouped together, as well as actions that can be matched. Previous research on this topic is based mainly on heuristics.

Large-margin structured estimation methods work by minimizing a convex upper bound of loss functions. While they allow for efficient optimization algorithms, these convex formulations are not tight and sacrifice the ability to accurately model the true loss. We present tighter non-convex bounds based on generalizing the notion of a ramp loss from binary classification to structured estimation. We show that a small modification of existing optimization algorithms suffices to solve this modified problem. On structured prediction tasks such as protein sequence alignment and web page ranking, our algorithm leads to improved accuracy.

Neurons perform computations, and convey the results of those computations through the statistical structure of their output spike trains. Here we present a practical method, grounded in the information-theoretic analysis of prediction, for inferring a minimal representation of that structure and for characterizing its complexity. Starting from spike trains, our approach finds their causal state models (CSMs), the minimal hidden Markov models or stochastic automata capable of generating statistically identical time series. We then use these CSMs to objectively quantify both the generalizable structure and the idiosyncratic randomness of the spike train. Specifically, we show that the expected algorithmic information content (the information needed to describe the spike train exactly) can be split into three parts describing (1) the time-invariant structure (complexity) of the minimal spike-generating process, which describes the spike train statistically; (2) the randomness (internal entropy rate) of the minimal spike-generating process; and (3) a residual pure noise term not described by the minimal spike-generating process. We use CSMs to approximate each of these quantities. The CSMs are inferred nonparametrically from the data, making only mild regularity assumptions, via the causal state splitting reconstruction algorithm. The methods presented here complement more traditional spike train analyses by describing not only spiking probability and spike train entropy, but also the complexity of a spike train's structure. We demonstrate our approach using both simulated spike trains and experimental data recorded in rat barrel cortex during vibrissa stimulation.