We compare and contrast the statistical physics and quantum physics inspired approaches for unsupervised generative modeling of classical data. The two approaches represent probabilities of observed data using energy-based models and quantum states respectively.Classical and quantum information patterns of the target datasets therefore provide principled guidelines for structural design and learning in these two approaches. Taking the restricted Boltzmann machines (RBM) as an example, we analyze the information theoretical bounds of the two approaches. We verify our reasonings by comparing the performance of RBMs of various architectures on the standard MNIST datasets.

Intelligent transportation systems (ITSs) will be a major component of tomorrow's smart cities. However, realizing the true potential of ITSs requires ultra-low latency and reliable data analytics solutions that can combine, in real-time, a heterogeneous mix of data stemming from the ITS network and its environment. Such data analytics capabilities cannot be provided by conventional cloud-centric data processing techniques whose communication and computing latency can be high. Instead, edge-centric solutions that are tailored to the unique ITS environment must be developed. In this paper, an edge analytics architecture for ITSs is introduced in which data is processed at the vehicle or roadside smart sensor level in order to overcome the ITS latency and reliability challenges. With a higher capability of passengers' mobile devices and intra-vehicle processors, such a distributed edge computing architecture can leverage deep learning techniques for reliable mobile sensing in ITSs. In this context, the ITS mobile edge analytics challenges pertaining to heterogeneous data, autonomous control, vehicular platoon control, and cyber-physical security are investigated. Then, different deep learning solutions for such challenges are proposed. The proposed deep learning solutions will enable ITS edge analytics by endowing the ITS devices with powerful computer vision and signal processing functions. Preliminary results show that the proposed edge analytics architecture, coupled with the power of deep learning algorithms, can provide a reliable, secure, and truly smart transportation environment.

Semi-supervised learning (SSL) partially circumvents the high cost of labeling data by augmenting a small labeled dataset with a large and relatively cheap unlabeled dataset drawn from the same distribution. This paper offers a novel interpretation of two deep learning-based SSL approaches, ladder networks and virtual adversarial training (VAT), as applying distributional smoothing to their respective latent spaces. We propose a class of models that fuse these approaches. We achieve near-supervised accuracy with high consistency on the MNIST dataset using just 5 labels per class: our best model, ladder with layer-wise virtual adversarial noise (LVAN-LW), achieves 1.42% +/- 0.12 average error rate on the MNIST test set, in comparison with 1.62% +/- 0.65 reported for the ladder network. On adversarial examples generated with L2-normalized fast gradient method, LVAN-LW trained with 5 examples per class achieves average error rate 2.4% +/- 0.3 compared to 68.6% +/- 6.5 for the ladder network and 9.9% +/- 7.5 for VAT.

Sparse coding is a crucial subroutine in algorithms for various signal processing, deep learning, and other machine learning applications. The central goal is to learn an overcomplete dictionary that can sparsely represent a given input dataset. However, a key challenge is that storage, transmission, and processing of the learned dictionary can be untenably high if the data dimension is high. In this paper, we consider the double-sparsity model introduced by Rubinstein et al. (2010b) where the dictionary itself is the product of a fixed, known basis and a data-adaptive sparse component. First, we introduce a simple algorithm for double-sparse coding that can be amenable to efficient implementation via neural architectures. Second, we theoretically analyze its performance and demonstrate asymptotic sample complexity and running time benefits over existing (provable) approaches for sparse coding. To our knowledge, our work introduces the first computationally efficient algorithm for double-sparse coding that enjoys rigorous statistical guarantees. Finally, we support our analysis via several numerical experiments on simulated data, confirming that our method can indeed be useful in problem sizes encountered in practical applications.

ABSTRACT A convolution neural network (CNN) based classification method for broadband DOA estimation is proposed, where the phase component of the short-time Fourier transform coefficients of the received microphone signals are directly fed into the CNN and the features required for DOA estimation are learned during training. Since only the phase component of the input is used, the CNN can be trained with synthesized noise signals, thereby making the preparation of the training data set easier compared to using speech signals. Through experimental evaluation, the ability of the proposed noise trained CNN framework to generalize to speech sources is demonstrated. In addition, the robustness of the system to noise, small perturbations in microphone positions, as well as its ability to adapt to different acoustic conditions is investigated using experiments with simulated and real data. Index Terms-- source localization, convolution neural networks, supervised learning, DOA estimation 1. INTRODUCTION Many applications such as hands-free communication, teleconferencing, and distant speech recognition require information on the location of a sound source in the acoustic environment.

We study two procedures (reverse-mode and forward-mode) for computing the gradient of the validation error with respect to the hyperparameters of any iterative learning algorithm such as stochastic gradient descent. These procedures mirror two methods of computing gradients for recurrent neural networks and have different trade-offs in terms of running time and space requirements. Our formulation of the reverse-mode procedure is linked to previous work by Maclaurin et al. [2015] but does not require reversible dynamics. The forward-mode procedure is suitable for real-time hyperparameter updates, which may significantly speed up hyperparameter optimization on large datasets. We present experiments on data cleaning and on learning task interactions. We also present one large-scale experiment where the use of previous gradient-based methods would be prohibitive.

Most modern supervised statistical/machine learning (ML) methods are explicitly designed to solve prediction problems very well. Achieving this goal does not imply that these methods automatically deliver good estimators of causal parameters. Examples of such parameters include individual regression coefficients, average treatment effects, average lifts, and demand or supply elasticities. In fact, estimates of such causal parameters obtained via naively plugging ML estimators into estimating equations for such parameters can behave very poorly due to the regularization bias. Fortunately, this regularization bias can be removed by solving auxiliary prediction problems via ML tools. Specifically, we can form an orthogonal score for the target low-dimensional parameter by combining auxiliary and main ML predictions. The score is then used to build a de-biased estimator of the target parameter which typically will converge at the fastest possible 1/root(n) rate and be approximately unbiased and normal, and from which valid confidence intervals for these parameters of interest may be constructed. The resulting method thus could be called a "double ML" method because it relies on estimating primary and auxiliary predictive models. In order to avoid overfitting, our construction also makes use of the K-fold sample splitting, which we call cross-fitting. This allows us to use a very broad set of ML predictive methods in solving the auxiliary and main prediction problems, such as random forest, lasso, ridge, deep neural nets, boosted trees, as well as various hybrids and aggregators of these methods.

AI has become part of the public consciousness. Researchers and data scientists have been sharing their groundbreaking work -- at what is officially known as the Conference and Workshop on Neural Information Processing Systems -- for three decades. But it's only with the recent explosion of interest in deep learning that NIPS has really taken off. We had two papers accepted to the conference this year, and contributed to two others. The researchers involved are among the 120 people on the NVIDIA Research team focused on pushing the boundaries of technology in machine learning, computer vision, self-driving cars, robotics, graphics, computer architecture, programming system, and other areas.

The most used methods are Regression, Clustering, Visualization, Decision Trees/Rules, and Random Forests; Deep Learning is used by only 20% of respondents; we also analyze which methods are most "industrial" and most "academic".

Don't you look at the CapsNet architecture and wonder... Wouldn't it have been amazing if I had come up with this idea? I mean, it was visible to all of us that pooling seemed just way too convenient amidst everything else about CNNs; just selecting the maximum weight among a specific number of weights and using that in the upcoming layers. Pooling was probably the easiest thing to visualize and understand in the entire architecture, which seemed very crude. But still, only the Godfather of Deep Learning did it again and came up with something brilliant -- adding layers inside existing layers instead of adding more layers i.e nested layers.... giving rise to the Capsule Networks! Improvements in CNNs started in the direction of adding more and more layers, playing with parameters and gradually towards connecting distant layers to each other to make sense out of their outputs once they were concatenated, when it was observed that simply increasing the number of layers also eventually reduces the performance after a certain point.