Interaction and Counterfactual Reasoning in Bayesian Decision Theory – We show how to apply the theory of objective reasoning to the contextual decision making task of evaluating two products from the same shopping cart, which we call product satisfaction in the context of objective logic. We provide an extension to the objective setting, and use this setting as the base for a new class of probabilistic knowledge-based decision making problems: the Decision-theoretic problem of decision making in online decision-making under uncertainty. In our proof, we provide a probabilistic interpretation of the problem and show how to use a probabilistic formal logic which we call objective calculus. We illustrate the theory and provide an example using a new problem of decision-making under uncertainty.

We consider the problem of learning sequential representations of data by leveraging sequential information for learning. In this paper, we establish a link between sequential and sequential knowledge via a connectionist framework using a novel set of constraints: given a dataset containing a subset of labels, an optimal sequence is selected by minimizing the minimum probability of all labels (i.e., the probability that the label is in the correct set). By combining the constraints with sequential knowledge, we infer sequential representations as a set of constraints. We show how this strategy, called sequential knowledge representation learning, can be extended to a set of more formal constraints and we show how to efficiently learn the sequential representations via sequential learning. We show how our approach can be used to guide downstream learning algorithms, such as classifiers, that use multiple constraints as a weight when learning. We provide theoretical and computational bounds on sequential knowledge representation learning and show how to use it to optimize a deep learning framework. Through experiments, we demonstrate that in some scenarios sequential knowledge representation learning helps reduce the computation cost of a sequential classification algorithm.

Object Recognition Using Adaptive Regularization

Directional Perception, Appearance, and Recognition

Interaction and Counterfactual Reasoning in Bayesian Decision Theory

The Power of Zero

Generative Closure Networks for Deep Neural NetworksWe consider the problem of learning sequential representations of data by leveraging sequential information for learning. In this paper, we establish a link between sequential and sequential knowledge via a connectionist framework using a novel set of constraints: given a dataset containing a subset of labels, an optimal sequence is selected by minimizing the minimum probability of all labels (i.e., the probability that the label is in the correct set). By combining the constraints with sequential knowledge, we infer sequential representations as a set of constraints. We show how this strategy, called sequential knowledge representation learning, can be extended to a set of more formal constraints and we show how to efficiently learn the sequential representations via sequential learning. We show how our approach can be used to guide downstream learning algorithms, such as classifiers, that use multiple constraints as a weight when learning. We provide theoretical and computational bounds on sequential knowledge representation learning and show how to use it to optimize a deep learning framework. Through experiments, we demonstrate that in some scenarios sequential knowledge representation learning helps reduce the computation cost of a sequential classification algorithm.