Can't pop-lock or Lindy Hop to save your life? Don't worry -- AI could soon make it look like you're a dance superstar. UC Berkeley researchers have developed a deep learning system that translates dance moves from a source video to less-than-experienced subjects. One algorithm creates a virtual skeleton to map poses, while two more algorithms square off against each other to both create the full picture and create a more realistic face for subjects as their virtual bodies twirl around. You do need the test subject to move around for a short while to get reference material, but the result is realistic enough to give an amateur the deftness of a ballet dancer.

Blockchain technology and artificial intelligence, two cutting-edge technologies, have the potential to change the face of healthcare as we know it by improving the quality and reducing costs through improved efficiencies. Most of us are at least somewhat familiar with artificial intelligence primarily through virtual assistants such as Siri and Alexa. Artificial intelligence automates repetitive learning and discovery through data after initially being set up by a human being. As many people also know, you have to be fairly specific when asking Siri and Alexa any questions -- the question must be posed in the right way -- to get the answer you are looking for. As an example, our interactions with Alexa, Siri, Google Search and Google Photos are based on deep learning.

This research presents a deep learning based approach to predict stress fields in the solid material elastic deformation using convolutional neural networks (CNN). Two different architectures are proposed to solve the problem. One is Feature Representation embedded Convolutional Neural Network (FR-CNN) with a single input channel, and the other is Squeeze-and-Excitation Residual network modules embedded Fully Convolutional Neural network (SE-Res-FCN) with multiple input channels. Both the tow architectures are stable and converged reliably in training and testing on GPUs. Accuracy analysis shows that SE-Res-FCN has a significantly smaller mean squared error (MSE) and mean absolute error (MAE) than FR-CNN. Mean relative error (MRE) of the SE-Res-FCN model is about 0.25% with respect to the average ground truth. The validation results indicate that the SE-Res-FCN model can accurately predict the stress field. For stress field prediction, the hierarchical architecture becomes deeper within certain limits, and then its prediction becomes more accurate. Fully trained deep learning models have higher computational efficiency over conventional FEM models, so they have great foreground and potential in structural design and topology optimization.

Finding meaningful representations in data is a core challenge in modern machine learning as the performance in many goal-directed frameworks such as reinforcement learning or supervised learning is directly and strongly influenced by the quality of the former. Usually, features are either provided (e.g. by expert knowledge) or acquired through learning. Most currently successful approaches for either deep supervised learning [1] or reinforcement learning [2] rely on a training signal, i.e., a classification label or reward signal, to provide sufficient indication which features of the input data should be extracted to increase performance. However, in most real-world scenarios labels have to be acquired by expert knowledge and reward signals are sparse. In unsupervised representation learning one tries to find and apply a principle by which to extract meaning from data without assuming the availability of any goal-driven metrics. Examples for such principles are based on reconstruction error (principal component analysis, autoencoder networks [3]), statistical dependence of extracted features 1 (independent component analysis [4]), indistinguishability of synthetically generated data from samples of the input distribution (generative adversarial nets [5]), fitting the probability distribution of input data (variational autoencoders [6]), (graph-)neighborhood preservation (locally-linear embedding [7], Laplacian eigenmaps [8]), and temporal coherence (slow feature analysis [9], regularized slowness optimization [10]).

Deep neural networks have been successfully deployed in a wide variety of applications including computer vision and speech recognition. However, computational and storage complexity of these models has forced the majority of computations to be performed on high-end computing platforms or on the cloud. To cope with computational and storage complexity of these models, this paper presents a training method that enables a radically different approach for realization of deep neural networks through Boolean logic minimization. The aforementioned realization completely removes the energy-hungry step of accessing memory for obtaining model parameters, consumes about two orders of magnitude fewer computing resources compared to realizations that use floatingpoint operations, and has a substantially lower latency.

Several recent papers investigate Active Learning (AL) for mitigating the data dependence of deep learning for natural language processing. However, the applicability of AL to real-world problems remains an open question. While in supervised learning, practitioners can try many different methods, evaluating each against a validation set before selecting a model, AL affords no such luxury. Over the course of one AL run, an agent annotates its dataset exhausting its labeling budget. Thus, given a new task, an active learner has no opportunity to compare models and acquisition functions. This paper provides a large scale empirical study of deep active learning, addressing multiple tasks and, for each, multiple datasets, multiple models, and a full suite of acquisition functions. We find that across all settings, Bayesian active learning by disagreement, using uncertainty estimates provided either by Dropout or Bayes-by Backprop significantly improves over i.i.d. baselines and usually outperforms classic uncertainty sampling.

This paper describes the system developed at Amobee for the WASSA 2018 implicit emotions shared task (IEST). The goal of this task was to predict the emotion expressed by missing words in tweets without an explicit mention of those words. We developed an ensemble system consisting of language models together with LSTM-based networks containing a CNN attention mechanism. Our approach represents a novel use of language models (specifically trained on a large Twitter dataset) to predict and classify emotions. Our system reached 1st place with a macro $\text{F}_1$ score of 0.7145.

Spatio-temporal problems are ubiquitous and of vital importance in many research fields. Despite the potential already demonstrated by deep learning methods in modeling spatio-temporal data, typical approaches tend to focus solely on conditional expectations of the output variables being modeled. In this paper, we propose a multi-output multi-quantile deep learning approach for jointly modeling several conditional quantiles together with the conditional expectation as a way to provide a more complete "picture" of the predictive density in spatio-temporal problems. Using two large-scale datasets from the transportation domain, we empirically demonstrate that, by approaching the quantile regression problem from a multi-task learning perspective, it is possible to solve the embarrassing quantile crossings problem, while simultaneously significantly outperforming state-of-the-art quantile regression methods. Moreover, we show that jointly modeling the mean and several conditional quantiles not only provides a rich description about the predictive density that can capture heteroscedastic properties at a neglectable computational overhead, but also leads to improved predictions of the conditional expectation due to the extra information and a regularization effect induced by the added quantiles.

Smart devices of everyday use (such as smartphones and wearables) are increasingly integrated with sensors that provide immense amounts of information about a person's daily life such as behavior and context. The automatic and unobtrusive sensing of behavioral context can help develop solutions for assisted living, fitness tracking, sleep monitoring, and several other fields. Towards addressing this issue, we raise the question: can a machine learn to recognize a diverse set of contexts and activities in a real-life through joint learning from raw multi-modal signals (e.g. accelerometer, gyroscope and audio etc.)? In this paper, we propose a multi-stream temporal convolutional network to address the problem of multi-label behavioral context recognition. A four-stream network architecture handles learning from each modality with a contextualization module which incorporates extracted representations to infer a user's context. Our empirical evaluation suggests that a deep convolutional network trained end-to-end achieves an optimal recognition rate. Furthermore, the presented architecture can be extended to include similar sensors for performance improvements and handles missing modalities through multi-task learning without any manual feature engineering on highly imbalanced and sparsely labeled dataset.

Deep learning is finding its way into the embedded world with applications such as autonomous driving, smart sensors and aug- mented reality. However, the computation of deep neural networks is demanding in energy, compute power and memory. Various approaches have been investigated to reduce the necessary resources, one of which is to leverage the sparsity occurring in deep neural networks due to the high levels of redundancy in the network parameters. It has been shown that sparsity can be promoted specifically and the achieved sparsity can be very high. But in many cases the methods are evaluated on rather small topologies. It is not clear if the results transfer onto deeper topologies. In this paper, the TensorQuant toolbox has been extended to offer a platform to investigate sparsity, especially in deeper models. Several practical relevant topologies for varying classification problem sizes are investigated to show the differences in sparsity for activations, weights and gradients.