Researchers used deep learning to create a new laser-based system that enables imaging around corners in real time. Hiding behind a wall might not be practical for much longer thanks to new technology that uses artificial intelligence to see and even read around corners. A team of researchers from Princeton, Stanford, Rice and Southern Methodist universities devised a system that uses powerful lasers similar to a laser pointer. The beam is bounced off a visible wall and onto a hidden object behind a corner. The beam then bounces off the object and back onto the wall.

The global deep learning market is expected to grow at a CAGR of 51.1% from forecast period 2019 to 2026 and expected to reach the value of around US$ 56,427.2 Deep learning is a subdivision of machine learning in artificial intelligence (AI) concerned with the algorithm inspired by the functioning of human brain termed as artificial neural networks. It is also termed as deep neural learning or deep neural network. Deep learning is evolved with the increasing amount of unstructured data due to digitalization. The available amount of data is utilized in deep learning to process or understand that data for effective decision making in various industry verticals including healthcare, manufacturing, automotive, agriculture, retail, security, human resources, marketing, law, and fintech.

Uber AI Labs has developed an algorithm called Generative Teaching Networks (GTN) that produces synthetic training data for neural networks which allows the networks to be trained faster than when using real data. Using this synthetic data, Uber sped up its neural architecture search (NAS) deep-learning optimization process by 9x. In a paper published on arXiv, the team described the system and a series of experiments. GTN is motivated by the problem of neural architecture search (NAS), which trains many different deep-learning model structures and selects the one that performs best on a set of test data. While a typical approach would train each model on the full data set for multiple iterations (or epochs), this is time-consuming and expensive.

Is it possible to make artificial intelligence more trustworthy by inserting a human being into the decision process of machine learning? It may be, but you don't get something for nothing. That human being better be an individual who knows a lot about what the neural network is trying to figure out. And that presents a conundrum, given that one of the main promises of AI is precisely to find out things humans don't know. It's a conundrum that is sidestepped in a new bit of AI work by scientists at the Technische Universität Darmstadt in Germany.

Pull stock prices from online API and perform predictions using Recurrent Neural Network & Long Short Term Memory (LSTM) with TensorFlow.js framework Machine learning is becoming increasingly popular these days and a growing number of the world's population see it is as a magic crystal ball: predicting when and what will happen in the future. This experiment uses artificial neural networks to reveal stock market trends and demonstrates the ability of time series forecasting to predict future stock prices based on past historical data. Disclaimer: As stock markets fluctuation are dynamic and unpredictable owing to multiple factors, this experiment is 100% educational and by no means a trading prediction tool. Before we can train the neural network and make any predictions, we will first require data. The type of data we are looking for is time series: a sequence of numbers in chronological order.

Despite the promising results of convolutional neural networks (CNNs), applying them on resource limited devices is still a challenge, mainly due to the huge memory and computation requirements. To tackle these problems, pruning can be applied to reduce the network size and number of floating point operations (FLOPs). Contrary to the \emph{filter norm} method -- that is used in network pruning and uses the assumption that the smaller this norm, the less important is the associated component --, we develop a novel filter importance norm that incorporates the loss caused by the elimination of a component from the CNN. To estimate the importance of a set of architectural components, we measure the CNN performance as different components are removed. The result is a collection of filter ensembles -- filter masks -- and associated performance values. We rank the filters based on a linear and additive model and remove the least important ones such that the drop in network accuracy is minimal. We evaluate our method on a fully connected network, as well as on the ResNet architecture trained on the CIFAR-10 data-set. Using our pruning method, we managed to remove $60\%$ of the parameters and $64\%$ of the FLOPs from the ResNet with an accuracy drop of less than $0.6\%$.

Continuous representation of words is a standard component in deep learning-based NLP models. However, representing a large vocabulary requires significant memory, which can cause problems, particularly on resource-constrained platforms. Therefore, in this paper we propose an isotropic iterative quantization (IIQ) approach for compressing embedding vectors into binary ones, leveraging the iterative quantization technique well established for image retrieval, while satisfying the desired isotropic property of PMI based models. Experiments with pre-trained embeddings (i.e., GloVe and HDC) demonstrate a more than thirty-fold compression ratio with comparable and sometimes even improved performance over the original real-valued embedding vectors.

Biological data are generally high-dimensional and require efficient machine learning methods that are well generalized and scalable to discover their complex nonlinear patterns. The recent advances in the domain of artificial intelligence and machine learning can be attributed to deep neural networks (DNNs) because they accomplish a variety of tasks in computer vision and natural language processing. However, standard DNNs are not suitable for handling high-dimensional data and data with small number of samples because they require a large pool of computing resources as well as plenty of samples to learn a large number of parameters. In particular, although interpretability is important for high-dimensional biological data such as gene expression data, a nonlinear feature selection algorithm for DNN models has not been fully investigated. In this paper, we propose a novel nonlinear feature selection method called the Feature Selection Network (FsNet), which is a scalable concrete neural network architecture, under high-dimensional and small number of samples setups. Specifically, our network consists of a selector layer that uses a concrete random variable for discrete feature selection and a supervised deep neural network regularized with the reconstruction loss. Because a large number of parameters in the selector and reconstruction layer can easily cause overfitting under a limited number of samples, we use two tiny networks to predict the large virtual weight matrices of the selector and reconstruction layers. The experimental results on several real-world high-dimensional biological datasets demonstrate the efficacy of the proposed approach.

We present a general numerical approach for learning unknown dynamical systems using deep neural networks (DNNs). Our method is built upon recent studies that identified the residue network (ResNet) as an effective neural network structure. In this paper, we present a generalized ResNet framework and broadly define residue as the discrepancy between observation data and prediction made by another model, which can be an existing coarse model or reduced-order model. In this case, the generalized ResNet serves as a model correction to the existing model and recovers the unresolved dynamics. When an existing coarse model is not available, we present numerical strategies for fast creation of coarse models, to be used in conjunction with the generalized ResNet. These coarse models are constructed using the same data set and thus do not require additional resources. The generalized ResNet is capable of learning the underlying unknown equations and producing predictions with accuracy higher than the standard ResNet structure. This is demonstrated via several numerical examples, including long-term prediction of a chaotic system.

Next point-of-interest (POI) recommendation aims to offer suggestions on which POI to visit next, given a user's POI visit history. This problem has a wide application in the tourism industry, and it is gaining an increasing interest as more POI check-in data become available. The problem is often modeled as a sequential recommendation problem to take advantage of the sequential patterns of user check-ins, e.g., people tend to visit Central Park after The Metropolitan Museum of Art in New York City. Recently, self-attentive networks have been shown to be both effective and efficient in general sequential recommendation problems, e.g., to recommend products, video games, or movies. Directly adopting self-attentive networks for next POI recommendation, however, may produce sub-optimal recommendations. This is because vanilla self-attentive networks do not consider the spatial and temporal patterns of user check-ins, which are two critical features in next POI recommendation. To address this limitation, in this paper, we propose a model named SANST that incorporates spatio-temporal patterns of user check-ins into self-attentive networks. To incorporate the spatial patterns, we encode the relative positions of POIs into their embeddings before feeding the embeddings into the self-attentive network. To incorporate the temporal patterns, we discretize the time of POI check-ins and model the temporal relationship between POI check-ins by a relation-aware self-attention module. We evaluate the performance of our SANST model with three real-world datasets. The results show that SANST consistently outperforms the state-of-theart models, and the advantage in nDCG@10 is up to 13.65%.