I consider the question "How should one act when the only goal is to learn as much as possible?". Building on the theoretical results of Fedorov (1972, Theory of Optimal Experiments, Academic Press) and MacKay (1992, Neural Computation, 4, 590-604), I apply techniques from optimal experiment design (OED) to guide the query/action selection of a neural network learner. I demonstrate that these t...

We describe an attentive encoder that combines tree-structured recursive neural networks and sequential recurrent neural networks for modelling sentence pairs. Since existing attentive models exert attention on the sequential structure, we propose a way to incorporate attention into the tree topology. Specially, given a pair of sentences, our attentive encoder uses the representation of one sen...

Interest in the effects of early adversity on children's development reflects contemporary emphases on early experience in the behavioral sciences and plasticity in the neurosciences. Over the past decade, powerful new tools and approaches for understanding the neural circuitry involved in emotion have become increasingly available. Yet, research in developmental psychopathology has not reaped ...

Recent research on the neural bases of decision making has focused on the role of the striatum and dopamine in learning to predict and obtain rewards. These studies have illuminated the processes by which repeated experiences of choice and reward feedback lead to the updating of value representations, allowing an organism to use past experience to improve choices when encountering the same situ...

Genetic screens in zebrafish have identified a large number of mutations that affect neural connectivity in the developing visual system. These mutants define genes essential for accurate retinal axon guidance in the eye and brain and the characterization of these mutants is helping to define the cellular and molecular mechanisms that guide axons in the vertebrate embryo. The combination of zeb...

Decision making involves selection from sets of options based on current evidence about the state of the world and estimates of the value or utility of different outcomes. The neural correlates of evidence assessment that guide simple perceptual judgments are now well understood, and we argue that this process can serve as a simple model for organizing behavioral and neural correlates of decisi...

Correction of atmospheric turbulences with the use of guide stars as reference, is one of the most relevant issues of adaptive optics (AO). This is addressed with tomographic techniques such as Multi-object adaptive optics (MOAO). Next generations of extremely large telescopes, will require improvements in computational capabilities of real time control systems. An improved version of CARMEN, a...

Several features for Neural Network based document region identification are tested. Specifically, this paper examines features for non-text region identification. The Neural Network based region identification algorithm is a key component of a document recognition system that segments a document into regions, classifies them into text, graphic, photo, and other region types, and then uses this...