Statistical characteristics of network traffic have attracted a significant amount of research for automated network intrusion detection, some of which looked at applications of natural statistical laws such as Zipf's law, Benford's law and the Pareto distribution. In this paper, we present the application of Benford's law to a new network flow metric "flow size difference", which have not been studied before by other researchers, to build an unsupervised flow-based intrusion detection system (IDS). The method was inspired by our observation on a large number of TCP flow datasets where normal flows tend to follow Benford's law closely but malicious flows tend to deviate significantly from it. The proposed IDS is unsupervised, so it can be easily deployed without any training. It has two simple operational parameters with a clear semantic meaning, allowing the IDS operator to set and adapt their values intuitively to adjust the overall performance of the IDS. We tested the proposed IDS on two (one closed and one public) datasets, and proved its efficiency in terms of AUC (area under the ROC curve). Our work showed the "flow size difference" has a great potential to improve the performance of any flow-based network IDSs.

Automatically searching for optimal hyperparameter configurations is of crucial importance for applying deep learning algorithms in practice. Recently, Bayesian optimization has been proposed for optimizing hyperparameters of various machine learning algorithms. Those methods adopt probabilistic surrogate models like Gaussian processes to approximate and minimize the validation error function of hyperparameter values. However, probabilistic surrogates require accurate estimates of sufficient statistics (e.g., covariance) of the error distribution and thus need many function evaluations with a sizeable number of hyperparameters. This makes them inefficient for optimizing hyperparameters of deep learning algorithms, which are highly expensive to evaluate. In this work, we propose a new deterministic and efficient hyperparameter optimization method that employs radial basis functions as error surrogates. The proposed mixed integer algorithm, called HORD, searches the surrogate for the most promising hyperparameter values through dynamic coordinate search and requires many fewer function evaluations. HORD does well in low dimensions but it is exceptionally better in higher dimensions. Extensive evaluations on MNIST and CIFAR-10 for four deep neural networks demonstrate HORD significantly outperforms the well-established Bayesian optimization methods such as GP, SMAC, and TPE. For instance, on average, HORD is more than 6 times faster than GP-EI in obtaining the best configuration of 19 hyperparameters.

Identification of falls while performing normal activities of daily living (ADL) is important to ensure personal safety and well-being. However, falling is a short term activity that occurs infrequently. This poses a challenge to traditional classification algorithms, because there may be very little training data for falls (or none at all). This paper proposes an approach for the identification of falls using a wearable device in the absence of training data for falls but with plentiful data for normal ADL. We propose three `X-Factor' Hidden Markov Model (XHMMs) approaches. The XHMMs model unseen falls using "inflated" output covariances (observation models). To estimate the inflated covariances, we propose a novel cross validation method to remove "outliers" from the normal ADL that serve as proxies for the unseen falls and allow learning the XHMMs using only normal activities. We tested the proposed XHMM approaches on two activity recognition datasets and show high detection rates for falls in the absence of fall-specific training data. We show that the traditional method of choosing a threshold based on maximum of negative of log-likelihood to identify unseen falls is ill-posed for this problem. We also show that supervised classification methods perform poorly when very limited fall data are available during the training phase.

Which authors of this paper are endorsers? Disable MathJax (What is MathJax?)

Decades of research have shown empathic communicators are more likely to have satisfying interactions and achieve their interactional goals, such as selling their products, delivering a better service experience or helping their patients. There's a popular notion empathic communication comes more naturally for some people than others. While it's true that there is individual variation, the assumption we cannot improve our ability to communicate empathically is incorrect. Through training and feedback we can significantly improve our ability to accurately interpret nonverbal behaviors. We can become better at reading other's nonverbal behaviors.

Harvard Wyss Institute researchers have been working on a soft exosuit with DARPA's financial help for years. While they were able to present a proof of concept in 2016, it's only now that they've found out just how much the suit can actually help its wearer. According to a new study published in Science Robotics, Harvard's exosuit reduces the energy a user needs to exert while walking by 23 percent. It does that by providing assistive force to the ankle at the perfect moment when you take another step. Team leader Conor Walsh said that's the highest percentage of reduction in energy use observed with an exosuit: "In a test group of seven healthy wearers, we clearly saw that the more assistance provided to the ankle joints, the more energy the wearers could save with a maximum reduction of almost 23% compared to walking with the exosuit powered-off. To our knowledge, this is the highest relative reduction in energy expenditure observed to date with a tethered exoskeleton or exosuit."

The healthcare headlines this year have been dominated by the imminent repeal of the Affordable Care Act (ACA). However, against the backdrop of a long-term transition to value-based care (VBC), a handful of emerging technology initiatives are quietly making news in advancing precision medicine in healthcare. The promise of precision medicine requires complete access to all available data about an individual. Over the past few years, digitization of health records through the implementation of EHR systems has covered the vast majority of hospitals and physician practices. Efforts to unlock value from unstructured data are already under way using natural language processing (NLP) technologies.

Ants are even more impressive at navigating than we thought. Scientists say they can follow a compass route, regardless of the direction in which they are facing. It is the equivalent of trying to find your way home while walking backwards or even spinning round and round. Experiments suggest ants keep to the right path by plotting the Sun's position in the sky which they combine with visual information about their surroundings. "Our main finding is that ants can decouple their direction of travel from their body orientation," said Dr Antoine Wystrach of the University of Edinburgh and CNRS in Paris.

Tesla has been cleared in a National Highway Traffic Safety Administration investigation over a crash involving a Tesla Model S last year, the department announced Thursday. In its report, the Traffic Safety Administration looked into the role that the Tesla's Autopilot feature played in the crash. Investigators did not find any major defects or issues with the car's autonomous features or support systems. "The Autopilot system is an Advanced Driver Assistance System (ADAS) that requires the continual and full attention of the driver to monitor the traffic environment and be prepared to take action to avoid crashes," the report said. "Tesla's design included hands-on [support to] the steering wheel system for monitoring driver engagement. That system has been updated to further reinforce the need for driver engagement through a "strike out" strategy. Drivers that do not respond to visual cues in the driver monitoring system alerts may "strike out" and lose Autopilot function for the remainder of the drive cycle."

Artificial intelligence can now outperform humans on a standard intelligence test. A new computational model scores within the 75th percentile, better than the average person, on a test known as Raven's Progressive Matrices. Researchers say this demonstrates that it can take on abstract visual reasoning tasks, and is a major step toward AI that can see and understand the world the way we do. Using Raven's Progressive Matrices, a nonverbal standardized test that measures abstract reasoning, the team found that their model is not only on par with humans, but performs better than many. In this example, participants choose which shape should come next in the sequence.