Recent works on gradient-based attacks and universal perturbations can adversarially modify images to bring down the accuracy of state-of-the-art classification techniques based on deep neural networks to as low as 10% on popular datasets like MNIST and ImageNet. The design of general defense strategies against a wide range of such attacks remains a challenging problem. In this paper, we derive inspiration from recent advances in the fields of cybersecurity and multi-agent systems and propose to use the concept of Moving Target Defense (MTD) for increasing the robustness of a set of deep networks against such adversarial attacks. To this end, we formalize and exploit the notion of differential immunity of an ensemble of networks to specific attacks. To classify an input image, a trained network is picked from this set of networks by formulating the interaction between a Defender (who hosts the classification networks) and their (Legitimate and Malicious) Users as a repeated Bayesian Stackelberg Game (BSG).We empirically show that our approach, MTDeep reduces misclassification on perturbed images for MNIST and ImageNet datasets while maintaining high classification accuracy on legitimate test images. Lastly, we demonstrate that our framework can be used in conjunction with any existing defense mechanism to provide more resilience to adversarial attacks than those defense mechanisms by themselves.

In this paper, we consider the problem of detecting unknown cyberattacks from audit data of system-level events. A key challenge is that different cyberattacks will have different suspicion indicators, which are not known beforehand. To address this we consider a multi-view anomaly detection framework, where multiple expert-designed ``views" of the data are created for capturing features that may serve as potential indicators. Anomaly detectors are then applied to each view and the results are combined to yield an overall suspiciousness ranking of system entities. Unfortunately, there is often a mismatch between what anomaly detection algorithms find and what is actually malicious, which can result in many false positives. This problem is made even worse in the multi-view setting, where only a small subset of the views may be relevant to detecting a particular cyberattack. To help reduce the false positive rate, a key contribution of this paper is to incorporate feedback from security analysts about whether proposed suspicious entities are of interest or likely benign. This feedback is incorporated into subsequent anomaly detection in order to improve the suspiciousness ranking toward entities that are truly of interest to the analyst. For this purpose, we propose an easy to implement variant of the perceptron learning algorithm, which is shown to be quite effective on benchmark datasets. We evaluate our overall approach on real attack data from a DARPA red team exercise, which include multiple attacks on multiple operating systems. The results show that the incorporation of feedback can significantly reduce the time required to identify malicious system entities.

Among the issues investigators on the U.S. Senate Intelligence Committee and Democrats on the House Intelligence Committee are digging into are whether IRA and other Russian organizations used any Facebook data, the sources said, speaking on condition of anonymity. Also, whether the use of such data had any impact on the U.S. election, and how much Facebook data may have been acquired by Russian entities, the sources said.

At 48 years old, Kersti Kaljulaid is Estonia's youngest president ever, and its first female president. A marathon runner with degrees in genetics and an MBA, she spent a career behind the scenes--mostly as a European government auditor--before being elected by Estonia's legislature in 2016. Known for its digital government, tax, and medical systems, Estonia is planning for the future. The country's "e-resident" program--which allows global citizens to obtain a government-issued ID card and set up remotely-operated businesses in Estonia--has attracted 35,000 people since 2014. Now the government is discussing a proposal to grant some rights to artificially intelligent systems.

AI research is increasingly being used by militaries around the world for offensive and defensive applications. This past week, groups of AI researchers began to fight back against two separate programs located halfway around the world from each other, generating tough questions about just how much engineers can affect the future uses of these technologies. From Silicon Valley, the New York Times published an internal protest memo written by several thousand Google employees, which vociferously opposed Google's work on a Defense Department-led initiative called Project Maven, which aims to use computer vision algorithms to analyze vast troves of image and video data. As the department's news service quoted Marine Corps Col. Drew Cukor last year about the initiative: "You don't buy AI like you buy ammunition," he added. "There's a deliberate workflow process and what the department has given us with its rapid acquisition authorities is an opportunity for about 36 months to explore what is governmental and [how] best to engage industry [to] advantage the taxpayer and the warfighter, who wants the best algorithms that exist to augment and complement the work he does."

In this special guest feature, Hal Lonas, Chief Technology Officer at Webroot, suggests that almost every software and information product and practice are thinking not "if," but "when and how" to apply AI and ML. If they don't, the competition will pass them up or make them irrelevant soon. There are many factors that have converged to enable this leap forward – let's look at some of them. Hal Lonas is CTO at Webroot, a privately held internet security company that provides state-of-the-art, cloud-based software as a service (SaaS) solutions spanning threat intelligence, detection and remediation. Previously the Senior VP of Product Engineering for Webroot, Lonas has 25 years of experience in enterprise software and engineering.

AI experts have warned that a South Korean University is in the process of developing a secret robot army that could destroy humanity. Top academics claim the Korea Advanced Institute of Science and Technology (Kaist) is working with weapons manufacturer Hanwha Systems to develop the technology. More than 50 leading academics from 30 different countries have now signed a letter boycotting the institution and expressing concern about its AI plans. Calling it a'Pandora's box', the experts believe AI and automated killing droids should not be used as weapons of war. Experts are distressed at the possibility of AI robots being developed for malicious purposes and claim it could lead to a third revolution in warfare.

Conversations about artificial intelligence are nothing new, nor are interviews with politicians about national policies. What is relatively new though, is hearing a national leader talk openly about a nation's relationship with (and policies regarding) artificial intelligence. Emmanuel Macron, France's president, recently gave an extensive interview to Wired magazine, where he spoke about France's goals with AI. President Macron said that the two applications of AI that made him realize how powerful, useful, and beneficial the technology could be, were applications in the healthcare and autonomous driving sectors. Macron says that he was impressed by new innovations that may allow healthcare to become much more personalized and allow doctors/researchers to better predict, diagnose, and treat illnesses.

New reports show that Google has ten times the personal information stored on you as Facebook. Tony Spitz has the details. More than 3,000 Google employees have signed a letter asking management end the company's involvement in Project Maven, a Defense Department drone surveillance project. The employees, in a letter addressed to company CEO Sundar Pichai, say Google's assistance in developing the artificial intelligence-powered system to detect vehicles and other objects in video captured by military drones betrays the company's motto of "Don't Be Evil." Google counters the employees' arguments saying, in a statement, the company's involvement is for "non-offensive purposes" and is used "to flag images for human review and is intended to save lives and save people from having to do highly tedious work."

It's hard to exaggerate just how successful NASA's Mars Rover program has been. These little vehicles have crawled over different parts of the Martian landscape, sending back invaluable data. But these rovers have some limitations: They move incredibly slowly. In over 2,000 days on Mars, the rover Curiosity has traveled about 11 and a half miles. That's why NASA has approved exploratory funding for an entirely new type of explorer: a swarm of robotic bees controlled by AI.