To what extent is the success of deep visualization due to the training? Could we do deep visualization using untrained, random weight networks? To address this issue, we explore new and powerful generative models for three popular deep visualization tasks using untrained, random weight convolutional neural networks. First we invert representations in feature spaces and reconstruct images from white noise inputs. The reconstruction quality is statistically higher than that of the same method applied on well trained networks with the same architecture. Next we synthesize textures using scaled correlations of representations in multiple layers and our results are almost indistinguishable with the original natural texture and the synthesized textures based on the trained network. Third, by recasting the content of an image in the style of various artworks, we create artistic images with high perceptual quality, highly competitive to the prior work of Gatys et al. on pretrained networks. To our knowledge this is the first demonstration of image representations using untrained deep neural networks. Our work provides a new and fascinating tool to study the representation of deep network architecture and sheds light on new understandings on deep visualization. It may possibly lead to a way to compare network architectures without training.

Spectral methods are popular in detecting global structures in the given data that can be represented as a matrix. However when the data matrix is sparse or noisy, classic spectral methods usually fail to work, due to localization of eigenvectors (or singular vectors) induced by the sparsity or noise. In this work, we propose a general method to solve the localization problem by learning a regularization matrix from the localized eigenvectors. Using matrix perturbation analysis, we demonstrate that the learned regularizations suppress down the eigenvalues associated with localized eigenvectors and enable us to recover the informative eigenvectors representing the global structure. We show applications of our method in several inference problems: community detection in networks, clustering from pairwise similarities, rank estimation and matrix completion problems. Using extensive experiments, we illustrate that our method solves the localization problem and works down to the theoretical detectability limits in different kinds of synthetic data. This is in contrast with existing spectral algorithms based on data matrix, non-backtracking matrix, Laplacians and those with rank-one regularizations, which perform poorly in the sparse case with noise.

In modern data analysis, random sampling is an efficient and widely-used strategy to overcome the computational difficulties brought by large sample size. In previous studies, researchers conducted random sampling which is according to the input data but independent on the response variable, however the response variable may also be informative for sampling. In this paper we propose an adaptive sampling called the gradient-based sampling which is dependent on both the input data and the output for fast solving of least-square (LS) problems. We draw the data points by random sampling from the full data according to their gradient values. This sampling is computationally saving, since the running time of computing the sampling probabilities is reduced to O(nd) where n is the full sample size and d is the dimension of the input. Theoretically, we establish an error bound analysis of the general importance sampling with respect to LS solution from full data. The result establishes an improved performance of the use of our gradient-based sampling. Synthetic and real data sets are used to empirically argue that the gradient-based sampling has an obvious advantage over existing sampling methods from two aspects of statistical efficiency and computational saving.

Deep convolutional neural networks (CNN) have achieved great success. On the other hand, modeling structural information has been proved critical in many vision problems. It is of great interest to integrate them effectively. In a classical neural network, there is no message passing between neurons in the same layer. In this paper, we propose a CRF-CNN framework which can simultaneously model structural information in both output and hidden feature layers in a probabilistic way, and it is applied to human pose estimation. A message passing scheme is proposed, so that in various layers each body joint receives messages from all the others in an efficient way. Such message passing can be implemented with convolution between features maps in the same layer, and it is also integrated with feedforward propagation in neural networks. Finally, a neural network implementation of end-to-end learning CRF-CNN is provided. Its effectiveness is demonstrated through experiments on two benchmark datasets.

This paper introduces an approach to regularize 2.5D surface normal and depth predictions at each pixel given a single input image. The approach infers and reasons about the underlying 3D planar surfaces depicted in the image to snap predicted normals and depths to inferred planar surfaces, all while maintaining fine detail within objects. Our approach comprises two components: (i) a fourstream convolutional neural network (CNN) where depths, surface normals, and likelihoods of planar region and planar boundary are predicted at each pixel, followed by (ii) a dense conditional random field (DCRF) that integrates the four predictions such that the normals and depths are compatible with each other and regularized by the planar region and planar boundary information. The DCRF is formulated such that gradients can be passed to the surface normal and depth CNNs via backpropagation. In addition, we propose new planar wise metrics to evaluate geometry consistency within planar surfaces, which are more tightly related to dependent 3D editing applications. We show that our regularization yields a 30% relative improvement in planar consistency on the NYU v2 dataset.

Existing object proposal algorithms usually search for possible object regions over multiple locations and scales \emph{ separately}, which ignore the interdependency among different objects and deviate from the human perception procedure. To incorporate global interdependency between objects into object localization, we propose an effective Tree-structured Reinforcement Learning (Tree-RL) approach to sequentially search for objects by fully exploiting both the current observation and historical search paths. The Tree-RL approach learns multiple searching policies through maximizing the long-term reward that reflects localization accuracies over all the objects. Starting with taking the entire image as a proposal, the Tree-RL approach allows the agent to sequentially discover multiple objects via a tree-structured traversing scheme. Allowing multiple near-optimal policies, Tree-RL offers more diversity in search paths and is able to find multiple objects with a single feed-forward pass. Therefore, Tree-RL can better cover different objects with various scales which is quite appealing in the context of object proposal. Experiments on PASCAL VOC 2007 and 2012 validate the effectiveness of the Tree-RL, which can achieve comparable recalls with current object proposal algorithms via much fewer candidate windows.

Foxconn, the Taiwanese manufacturing giant behind Apple's iPhone and numerous other major electronics devices, aims to automate away a vast majority of its human employees, according to a report from DigiTimes. Dai Jia-peng, the general manager of Foxconn's automation committee, says the company has a three-phase plan in place to automate its Chinese factories using software and in-house robotics units, known as Foxbots. The first phase of Foxconn's automation plans involve replacing the work that is either dangerous or involves repetitious labor humans are unwilling to do. The second phase involves improving efficiency by streamlining production lines to reduce the number of excess robots in use. The third and final phase involves automating entire factories, "with only a minimal number of workers assigned for production, logistics, testing, and inspection processes," according to Jia-peng.

Artificial intelligence (AI) made incredible strides in 2016, and the growth appears set to accelerate as we enter the New Year. A team of Microsoft researchers has released a dataset of 100,000 questions and answers that other AI researchers can use – for free – in their quest to create systems that can read and answer questions as well as a human. The MS MARCO dataset is based on anonymized real-world data from Bing and Cortana queries and is part of an attempt to spur the breakthroughs in machine reading that are already happening in image and speech recognition. The move is also aimed at facilitating advances toward "artificial general intelligence," or machines that can think like humans – and can read and understand a document as well as a person. Meanwhile, AI helped a musician in Norway sing a new tune for the holidays this year: a Christmas carol that was created by Microsoft's AI technology.

You can add Boston to the list of places where autonomous vehicles are being tested legally. Rather than trials on the city's, ahem, interesting street layout, company NuTonomy will start small, beginning at the Raymond L. Flynn Marine Park on January 3rd, according to The Boston Herald. If you'll remember, nuTonomy is operating a self-driving taxi service in Singapore. The test vehicle is a Renault Zoe -- one of the models it uses for autonomous taxis -- but that's where the similarities end. The Herald says there aren't any plans to offer a similar service in the city just yet.

Marketing automation has been around since the late 1990s. However, artificial intelligence has only recently begun to realize its full potential. How will new AI technology transform the marketing profession over the next few years? Here are a few things to look out for. Almost all brands understand the importance of customer satisfaction.