The paper reviews an emerging body of theoretical results on deep learning including the conditions under which it can be exponentially better than shallow learning. Deep convolutional networks represent an important special case of these conditions, though weight sharing is not the main reason for their exponential advantage. Explanation of a few key theorems is provided together with new resu...

Juvenile common carp Cyprinus carpio were collected from 10 lakes with variable predator abundance over 4 months to evaluate if morphological defences increased with increasing predation risk. Cyprinus carpio dorsal and pectoral spines were longer and body depth was deeper when predators were more abundant, with differences becoming more pronounced from July to October. To determine if morpholo...

INTRODUCTION The orthographic depth hypothesis (Katz and Feldman, 1983) posits that different reading routes are engaged depending on the type of grapheme/phoneme correspondence of the language being read. Shallow orthographies with consistent grapheme/phoneme correspondences favor encoding via non-lexical pathways, where each grapheme is sequentially mapped to its corresponding phoneme. In con...

In the psycholinguistic literature it has been proposed that readers and listeners often adopt a "good-enough" processing strategy in which a "shallow" representation of an utterance driven by (top-down) extra-grammatical processes has a processing advantage over a "deep" (bottom-up) grammatically-driven representation of that same utterance. In the current contribution we claim, both on theore...

The increase in extreme heat events associated to global warming threatens seagrass ecosystems, likely by affecting key plant physiological processes such as photosynthesis and respiration. Understanding species' ability to acclimate to warming is crucial to better predict their future trends. Here, we study tolerance to warming in two key Mediterranean seagrasses, Posidonia oceanica and Cymodo...

The Monte Carlo Pseudo Inference Engine for Text (MCPIET) addresses the RTE problem within a new theoretic framework for robust inference and logical pattern processing based on integrated deep and shallow semantics. In this report we outline, in some detail, this new theoretic framework, and we will use it to shed some light on the informativity and robustness characteristics for the extreme c...

The present paper presents a simple theory for the transformation of nonprecipitating, shallow convection into precipitating, deep convective clouds. Tomake the pertinent point amuch idealized system is considered, consisting only of shallow and deep convection without large-scale forcing. The transformation is described by an explicit coupling between these two types of convection. Shallow con...

Ecosystem-level effects of stressors are critical to understanding community regulation, and environmental stress models are useful in describing such effects. Hypoxia is an important stressor in aquatic ecosystems that usually decreases abundance and biomass of benthic fauna. In field surveys, predator abundance is low in hypoxic areas, and in lab experiments, predators reduce their feeding ra...

To further characterize changes in functional brain development that are associated with the emergence of cognitive control, participants 14 to 28 years of age were scanned while performing an episodic encoding task with a levels-of-processing manipulation. Using data from the 12 youngest and oldest participants (endpoint groups), 18 regions were identified that showed group differences in task...

In this paper, we introduce transformations of deep rectifier networks, enabling the conversion of deep rectifier networks into shallow rectifier networks. We subsequently prove that any rectifier net of any depth can be represented by a maximum of a number of functions that can be realized by a shallow network with a single hidden layer. The transformations of both deep rectifier nets and deep...