We'll look at some of the most important papers that have been published over the last 5 years and discuss why they're so important. The first half of the list (AlexNet to ResNet) deals with advancements in general network architecture, while the second half is just a collection of interesting papers in other subareas. The one that started it all (Though some may say that Yann LeCun's paper in 1998 was the real pioneering publication). This paper, titled "ImageNet Classification with Deep Convolutional Networks", has been cited a total of 6,184 times and is widely regarded as one of the most influential publications in the field. Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton created a "large, deep convolutional neural network" that was used to win the 2012 ILSVRC (ImageNet Large-Scale Visual Recognition Challenge). For those that aren't familiar, this competition can be thought of as the annual Olympics of computer vision, where teams from across the world compete to see who has the best computer vision model for tasks such as classification, localization, detection, and more. The next best entry achieved an error of 26.2%, which was an astounding improvement that pretty much shocked the computer vision community. Safe to say, CNNs became household names in the competition from then on out. In the paper, the group discussed the architecture of the network (which was called AlexNet).

Read this introductory list of contemporary machine learning algorithms of importance that every engineer should understand. It is no doubt that the sub-field of machine learning / artificial intelligence has increasingly gained more popularity in the past couple of years. As Big Data is the hottest trend in the tech industry at the moment, machine learning is incredibly powerful to make predictions or calculated suggestions based on large amounts of data. Some of the most common examples of machine learning are Netflix's algorithms to make movie suggestions based on movies you have watched in the past or Amazon's algorithms that recommend books based on books you have bought before. So if you want to learn more about machine learning, how do you start?

Zero-Shot learning has been shown to be an efficient strategy for domain adaptation. In this context, this paper builds on the recent work of Bucher et al. [1], which proposed an approach to solve Zero-Shot classification problems (ZSC) by introducing a novel metric learning based objective function. This objective function allows to learn an optimal embedding of the attributes jointly with a measure of similarity between images and attributes. This paper extends their approach by proposing several schemes to control the generation of the negative pairs, resulting in a significant improvement of the performance and giving above state-of-the-art results on three challenging ZSC datasets.

The unscented transformation (UT) is an efficient method to solve the state estimation problem for a non-linear dynamic system, utilizing a derivative-free higher-order approximation by approximating a Gaussian distribution rather than approximating a non-linear function. Applying the UT to a Kalman filter type estimator leads to the well-known unscented Kalman filter (UKF). Although the UKF works very well in Gaussian noises, its performance may deteriorate significantly when the noises are non-Gaussian, especially when the system is disturbed by some heavy-tailed impulsive noises. To improve the robustness of the UKF against impulsive noises, a new filter for nonlinear systems is proposed in this work, namely the maximum correntropy unscented filter (MCUF). In MCUF, the UT is applied to obtain the prior estimates of the state and covariance matrix, and a robust statistical linearization regression based on the maximum correntropy criterion (MCC) is then used to obtain the posterior estimates of the state and covariance. The satisfying performance of the new algorithm is confirmed by two illustrative examples.

The amount of digitally available but heterogeneous information about the world is remarkable, and new technologies such as self-driving cars, smart homes, or the internet of things may further increase it. In this paper we examine certain aspects of the problem of how such heterogeneous information can be harnessed by autonomous agents. After discussing potentials and limitations of some existing approaches, we investigate how \emph{experiments} can help to obtain a better understanding of the problem. Specifically, we present a simple agent that integrates video data from a different agent, and implement and evaluate a version of it on the novel experimentation platform \emph{Malmo}. The focus of a second investigation is on how information about the hardware of different agents, the agents' sensory data, and \emph{causal} information can be utilized for knowledge transfer between agents and subsequently more data-efficient decision making. Finally, we discuss potential future steps w.r.t.\ theory and experimentation, and formulate open questions.

We present a unified approach for learning the parameters of Sum-Product networks (SPNs). We prove that any complete and decomposable SPN is equivalent to a mixture of trees where each tree corresponds to a product of univariate distributions. Based on the mixture model perspective, we characterize the objective function when learning SPNs based on the maximum likelihood estimation (MLE) principle and show that the optimization problem can be formulated as a signomial program. We construct two parameter learning algorithms for SPNs by using sequential monomial approximations (SMA) and the concave-convex procedure (CCCP), respectively. The two proposed methods naturally admit multiplicative updates, hence effectively avoiding the projection operation. With the help of the unified framework, we also show that, in the case of SPNs, CCCP leads to the same algorithm as Expectation Maximization (EM) despite the fact that they are different in general.

Robots have been illustrated as humans' mechanical servants, but experts are determined to turn these cyborgs into emotional synthetic beings. Now, researchers brought the two of the world's most advanced robots together to test their reactions by showing them the trailer for the horror flick'Morgan'. Edi vocalizes its fear with phrases such as'Oh no, I can't watch' and although FACE is silent, it offers its'thoughts' by eerily moving its eyes, mouth and head. Edi (Electronic Deceptive Intelligence) is the brainchild of magicLab.ny, Edi is a fitted with a range sensors, has long robotic arms and a screen that displays a cartoon face.

A dream is successions of images, ideas, emotions, and sensations that usually occurs involuntarily in the mind during certain stages of sleep.[1] The content and purpose of dreams are not definitively understood, though they have been a topic of scientific speculation, as well as a subject of philosophical and religious interest, throughout recorded history. The scientific study of dreams is called oneirology.[2] Dreams mainly occur in the rapid-eye movement (REM) stage of sleep--when brain activity is high and resembles that of being awake. REM sleep is revealed by continuous movements of the eyes during sleep. At times, dreams may occur during other stages of sleep. However, these dreams tend to be much less vivid or memorable.[3] The length of a dream can vary; they may last for a few seconds, or approximately 20–30 minutes.[3] People are more likely to remember the dream if they are awakened during the REM phase. The average person has three to five dreams per night, and some may have up to seven;[4] however, most dreams are immediately or quickly forgotten.[5] Dreams tend to last longer as the night progresses. During a full eight-hour night sleep, most dreams occur in the typical two hours of REM.[6] In modern times, dreams have been seen as a connection to the unconscious mind. They range from normal and ordinary to overly surreal and bizarre. Dreams can have varying natures, such as being frightening, exciting, magical, melancholic, adventurous, or sexual. The events in dreams are generally outside the control of the dreamer, with the exception of lucid dreaming, where the dreamer is self-aware.[7]

As part of our look back over the last 70 years, we created this quiz all about British motoring from 1946 to 2016.But enough about the past – we've always got one eye on the future, with exciting new developments happening in the automotive industry all the time. Right now, electric cars are growing in popularity, in-car infotainment tech is getting more and more impressive, and manufacturers are working with governments to make self-driving cars a reality. Within the next decade, Britain's roads could begin looking very different. In the year 2086, how might the motoring world have changed? There will be some major changes over the next 70 years.

WORLD superpowers are engaged in a feverish "arms race" to develop the first killer robots completely removed from human control, the Sun Online can reveal. These machines will mark a dramatic escalation in computer AI from the drones and robots currently in use, all of which still require a human to press the "kill button". In a series of exclusive interviews, leading experts told The Sun Online machines making life or death decisions will likely be developed within the next 10 years. Fears are now growing about the implications of creating such smart machines, as are concerns they will fall into the hands of terrorist groups such as ISIS. Locked in this new race for military supremacy is Britain, the US, China, Russia and Israel – all of which have robot programmes of varying advancement.