The present study proposes a deep learning model, named DeepSleepNet, for automatic sleep stage scoring based on raw single-channel EEG. Most of the existing methods rely on hand-engineered features which require prior knowledge of sleep analysis. Only a few of them encode the temporal information such as transition rules, which is important for identifying the next sleep stages, into the extracted features. In the proposed model, we utilize Convolutional Neural Networks to extract time-invariant features, and bidirectional-Long Short-Term Memory to learn transition rules among sleep stages automatically from EEG epochs. We implement a two-step training algorithm to train our model efficiently. We evaluated our model using different single-channel EEGs (F4-EOG(Left), Fpz-Cz and Pz-Oz) from two public sleep datasets, that have different properties (e.g., sampling rate) and scoring standards (AASM and R&K). The results showed that our model achieved similar overall accuracy and macro F1-score (MASS: 86.2%-81.7, Sleep-EDF: 82.0%-76.9) compared to the state-of-the-art methods (MASS: 85.9%-80.5, Sleep-EDF: 78.9%-73.7) on both datasets. This demonstrated that, without changing the model architecture and the training algorithm, our model could automatically learn features for sleep stage scoring from different raw single-channel EEGs from different datasets without utilizing any hand-engineered features.

Especially in the community of Digital Humanities, the automated processing of Latin texts has always been a popular research topic. In a variety of computational applications, such as text reuse detection [Franzini et al, 2015], it is desirable to annotate and augment Latin texts with useful morpho-syntactical or lexical information, such as lemmas. In this paper, we will focus on two sequence tagging tasks for medieval Latin: part-of-speech tagging and lemmatization. Given a piece of Latin text, the task of lemmatization involves assigning each word to a single dictionary headword or'lemma': a baseform label (preferably in a normalized orthography) grouping all word tokens which only differ in spelling and/or inflection [Knowles et al, 2004]. The task of lemmatization is closely related to that of part-of-speech (PoS) tagging [Jurafsky et al, 2000], in which each word in a running text should be assigned a tag indicating its part of speech or word class (e.g.

We show how the success of deep learning could depend not only on mathematics but also on physics: although well-known mathematical theorems guarantee that neural networks can approximate arbitrary functions well, the class of functions of practical interest can frequently be approximated through "cheap learning" with exponentially fewer parameters than generic ones. We explore how properties frequently encountered in physics such as symmetry, locality, compositionality, and polynomial log-probability translate into exceptionally simple neural networks. We further argue that when the statistical process generating the data is of a certain hierarchical form prevalent in physics and machine-learning, a deep neural network can be more efficient than a shallow one. We formalize these claims using information theory and discuss the relation to the renormalization group. We prove various "no-flattening theorems" showing when efficient linear deep networks cannot be accurately approximated by shallow ones without efficiency loss, for example, we show that $n$ variables cannot be multiplied using fewer than 2^n neurons in a single hidden layer.

Non-invasive detection of cardiovascular disorders from radiology scans requires quantitative image analysis of the heart and its substructures. There are well-established measurements that radiologists use for diseases assessment such as ejection fraction, volume of four chambers, and myocardium mass. These measurements are derived as outcomes of precise segmentation of the heart and its substructures. The aim of this paper is to provide such measurements through an accurate image segmentation algorithm that automatically delineates seven substructures of the heart from MRI and/or CT scans. Our proposed method is based on multi-planar deep convolutional neural networks (CNN) with an adaptive fusion strategy where we automatically utilize complementary information from different planes of the 3D scans for improved delineations. For CT and MRI, we have separately designed three CNNs (the same architectural configuration) for three planes, and have trained the networks from scratch for voxel-wise labeling for the following cardiac structures: myocardium of left ventricle (Myo), left atrium (LA), left ventricle (LV), right atrium (RA), right ventricle (RV), ascending aorta (Ao), and main pulmonary artery (PA). We have evaluated the proposed method with 4-fold-cross validation on the multi-modality whole heart segmentation challenge (MM-WHS 2017) dataset. The precision and dice index of 0.93 and 0.90, and 0.87 and 0.85 were achieved for CT and MR images, respectively. While a CT volume was segmented about 50 seconds, an MRI scan was segmented around 17 seconds with the GPUs/CUDA implementation.

Error bars represent the mean /- ‐ SEM across cross- ‐validation folds. X's represent the R 2 values of each cross- ‐validation fold. Note the different y- ‐axis limits for the hippocampus dataset. Figure 4: Example results with limited training data Using only 2 minutes of training data for motor cortex and somatosensory cortex, and 15 minutes of training data for hippocampus, we trained two traditional methods (Wiener filter and Kalman filter), and two modern methods (feedforward neural network and LSTM). Example decoding results are shown from motor cortex (left), somatosensory cortex (middle), and hippocampus (right), for these methods (top to bottom). Ground truth traces are in black, while decoder results are in the same colors as previous figures. Figure 5: Decoder results with varying amounts of training data Using varying amounts of training data, we trained two traditional methods (Wiener filter and Kalman filter), and two modern methods (feedforward neural network and LSTM). R 2 values are reported for these decoders (different colors) for each brain area (top to bottom).

This paper presents a Neural Aggregation Network (NAN) for video face recognition. The network takes a face video or face image set of a person with a variable number of face images as its input, and produces a compact, fixed-dimension feature representation for recognition. The whole network is composed of two modules. The feature embedding module is a deep Convolutional Neural Network (CNN) which maps each face image to a feature vector. The aggregation module consists of two attention blocks which adaptively aggregate the feature vectors to form a single feature inside the convex hull spanned by them. Due to the attention mechanism, the aggregation is invariant to the image order. Our NAN is trained with a standard classification or verification loss without any extra supervision signal, and we found that it automatically learns to advocate high-quality face images while repelling low-quality ones such as blurred, occluded and improperly exposed faces. The experiments on IJB-A, YouTube Face, Celebrity-1000 video face recognition benchmarks show that it consistently outperforms naive aggregation methods and achieves the state-of-the-art accuracy.

Correct inference of genetic regulations inside a cell is one of the greatest challenges in post genomic era for the biologist and researchers. Several intelligent techniques and models were already proposed to identify the regulatory relations among genes from the biological database like time series microarray data. Recurrent Neural Network (RNN) is one of the most popular and simple approach to model the dynamics as well as to infer correct dependencies among genes. In this paper, Bat Algorithm (BA) is applied to optimize the model parameters of RNN model of Gene Regulatory Network (GRN). Initially the proposed method is tested against small artificial network without any noise and the efficiency is observed in term of number of iteration, number of population and BA optimization parameters. The model is also validated in presence of different level of random noise for the small artificial network and that proved its ability to infer the correct inferences in presence of noise like real world dataset. In the next phase of this research, BA based RNN is applied to real world benchmark time series microarray dataset of E. coli. The results prove that it can able to identify the maximum number of true positive regulation but also include some false positive regulations. Therefore, BA is very suitable for identifying biological plausible GRN with the help RNN model.

Check out the Strata Business Summit at the Strata Data Conference in New York City, Sept. 25-28, 2017, to learn more from data-driven businesses--including American Express, BBC Worldwide, and LinkedIn. Early price ends August 11. Driverless cars aren't the only application for deep learning on the road: neural networks have begun to make their way into every corner of the automotive industry, from supply-chain management to engine controllers. In this installment of our ongoing series on artificial intelligence (AI) and machine learning (ML) in the enterprise, we speak with Dimitar Filev, executive technical leader at Ford Research & Advanced Engineering, who leads the team focused on control methods and computational intelligence. Ford research lab has been conducting systematic research on computational intelligence--one of the branches of AI--for more than 20 years.

Now Warden's tackling a new challenge: teaching smartphones (and cameras in general) how to recognize objects. Back in April, Warden released a software development kit called DeepBeliefSDK on GitHub. Designed for developers to integrate machine vision into smartphone apps, DeepBelief is currently available for Android, iOS, Linux, and Raspberry Pi. While DeepBelief is just one in a number of early entrants into the somewhat creepy world of deep learning for mobile devices, it has one advantage over competitors. It's blazing fast–Warden says Deep Belief can identify objects in under 300 milliseconds on an iPhone 5S, while using less than 20 megabytes of memory.

FILE - This Monday, Jan. 4, 2016, file photo, shows a pair of Under Armour SpeedForm Gemini 2 Record Equipped running shoes, containing an embedded chip to track exercise, on display, in New York. On Tuesday, Aug. 1, 2017, Under Armour announced it is cutting approximately 280 jobs from its global workforce and lowering its full-year revenue outlook, overshadowing a second-quarter performance that topped most expectations.