There hasn't been a major breakthrough in algorithms in the last year.

Large-scale labeled training datasets have enabled deep neural networks to excel on a wide range of benchmark vision tasks. However, in many applications it is prohibitively expensive or time-consuming to obtain large quantities of labeled data. To cope with limited labeled training data, many have attempted to directly apply models trained on a large-scale labeled source domain to another sparsely labeled target domain. Unfortunately, direct transfer across domains often performs poorly due to domain shift and dataset bias. Domain adaptation is the machine learning paradigm that aims to learn a model from a source domain that can perform well on a different (but related) target domain. In this paper, we summarize and compare the latest unsupervised domain adaptation methods in computer vision applications. We classify the non-deep approaches into sample re-weighting and intermediate subspace transformation categories, while the deep strategy includes discrepancy-based methods, adversarial generative models, adversarial discriminative models and reconstruction-based methods. We also discuss some potential directions.

Spike-and-Slab Deep Learning (SS-DL) is a fully Bayesian alternative to Dropout for improving generalizability of deep ReLU networks. This new type of regularization enables provable recovery of smooth input-output maps with unknown levels of smoothness. Indeed, we show that the posterior distribution concentrates at the near minimax rate for $\alpha$-H\"older smooth maps, performing as well as if we knew the smoothness level $\alpha$ ahead of time. Our result sheds light on architecture design for deep neural networks, namely the choice of depth, width and sparsity level. These network attributes typically depend on unknown smoothness in order to be optimal. We obviate this constraint with the fully Bayes construction. As an aside, we show that SS-DL does not overfit in the sense that the posterior concentrates on smaller networks with fewer (up to the optimal number of) nodes and links. Our results provide new theoretical justifications for deep ReLU networks from a Bayesian point of view.

Our aim is to study gradient descent in such loss functions or energy landscapes and compare it to results obtained from real high-dimensional optimization problems such as encountered in deep learning. In particular, we analyze the distribution of the improved loss function after a step of gradient descent, provide analytic expressions for the moments as well as prove asymptotic normality as the dimension of the parameter space becomes large. Moreover, we compare this with the expectation of the global minimum of the landscape obtained by means of the Euler characteristic of excursion sets. Besides complementing our analytical findings with numerical results from simulated Gaussian random fields, we also compare it to loss functions obtained from optimisation problems on synthetic and real data sets by proposing a "black box" random field toy-model for a deep neural network loss function.

Generative adversarial networks (GANs) have demonstrated to be successful at generating realistic real-world images. In this paper we compare various GAN techniques, both supervised and unsupervised. The effects on training stability of different objective functions are compared. We add an encoder to the network, making it possible to encode images to the latent space of the GAN. The generator, discriminator and encoder are parameterized by deep convolutional neural networks. For the discriminator network we experimented with using the novel Capsule Network, a state-of-the-art technique for detecting global features in images. Experiments are performed using a digit and face dataset, with various visualizations illustrating the results. The results show that using the encoder network it is possible to reconstruct images. With the conditional GAN we can alter visual attributes of generated or encoded images. The experiments with the Capsule Network as discriminator result in generated images of a lower quality, compared to a standard convolutional neural network.

Training deep neural networks (DNNs) efficiently is a challenge due to the associated highly nonconvex optimization. The backpropagation (backprop) algorithm has long been the most widely used algorithm for gradient computation of parameters of DNNs and is used along with gradient descent-type algorithms for this optimization task. Recent work have shown the efficiency of block coordinate descent (BCD) type methods empirically for training DNNs. In view of this, we propose a novel algorithm based on the BCD method for training DNNs and provide its global convergence results built upon the powerful framework of the Kurdyka-Lojasiewicz (KL) property. Numerical experiments on standard datasets demonstrate its competitive efficiency against standard optimizers with backprop.

We study the error landscape of deep linear and nonlinear neural networks with the squared error loss. Minimizing the loss of a deep linear neural network is a nonconvex problem, and despite recent progress, our understanding of this loss surface is still incomplete. For deep linear networks, we present necessary and sufficient conditions for a critical point of the risk function to be a global minimum. Surprisingly, our conditions provide an efficiently checkable test for global optimality, while such tests are typically intractable in nonconvex optimization. We further extend these results to deep nonlinear neural networks and prove similar sufficient conditions for global optimality, albeit in a more limited function space setting.

Hurricanes are cyclones circulating about a defined center whose closed wind speeds exceed 75 mph originating over tropical and subtropical waters. At landfall, hurricanes can result in severe disasters. The accuracy of predicting their trajectory paths is critical to reduce economic loss and save human lives. Given the complexity and nonlinearity of weather data, a recurrent neural network (RNN) could be beneficial in modeling hurricane behavior. We propose the application of a fully connected RNN to predict the trajectory of hurricanes. We employed the RNN over a fine grid to reduce typical truncation errors. We utilized their latitude, longitude, wind speed, and pressure publicly provided by the National Hurricane Center (NOAA) to predict the trajectory of a hurricane at 6-hour intervals.

User interfaces provide an interactive window between physical and virtual environments. A new concept in the field of human-computer interaction is a soft user interface; a compliant surface that facilitates touch interaction through deformation. Despite the potential of these interfaces, they currently lack a signal processing framework that can efficiently extract information from their deformation. Here we present OrbTouch, a device that uses statistical learning algorithms, based on convolutional neural networks, to map deformations from human touch to categorical labels (i.e., gestures) and touch location using stretchable capacitor signals as inputs. We demonstrate this approach by using the device to control the popular game Tetris. OrbTouch provides a modular, robust framework to interpret deformation in soft media, laying a foundation for new modes of human computer interaction through shape changing solids.

Artificial Intelligence is on a rage! All of a sudden every one, whether understands or not, is talking about it. Understanding the latest advancements in artificial intelligence can seem overwhelming, but it really boils down to two very popular concepts Machine Learning and Deep Learning. But lately, Deep Learning is gaining much popularity due to it's supremacy in terms of accuracy when trained with huge amount of data. Machine Learning has become necessary in every sector as a way of making machines intelligent. In a simpler way, Machine Learning is set of algorithms that parse data, learn from them, and then apply what they've learned to make intelligent decisions.