We identify obfuscated gradients, a kind of gradient masking, as a phenomenon that leads to a false sense of security in defenses against adversarial examples. While defenses that cause obfuscated gradients appear to defeat iterative optimization-based attacks, we find defenses relying on this effect can be circumvented. For each of the three types of obfuscated gradients we discover, we describe characteristic behaviors of defenses exhibiting this effect and develop attack techniques to overcome it. In a case study, examining non-certified white-box-secure defenses at ICLR 2018, we find obfuscated gradients are a common occurrence, with 7 of 8 defenses relying on obfuscated gradients. Our new attacks successfully circumvent 6 completely and 1 partially.

A NASA spacecraft on its way to investigate a nearby asteroid has caught a remarkable glimpse at the orbital dance of Earth and the moon from far away. In a new image shared today by the space agency, Earth and its only natural satellite are the two brightest specks in a black sky. The Osiris-Rex spacecraft snapped the incredible photo from 39.5 million miles away, as it flew by at 19,000 miles per hour (8.5 kilometers per second). A NASA spacecraft on its way to investigate a nearby asteroid has caught a remarkable glimpse at the orbital dance of Earth and the moon from far away. Osiris is on its way to the near-Earth asteroid'Bennu,' in a mission to collect a 2.1-ounce sample and return it to Earth for study.

If you make a purchase by clicking one of our links, we may earn a small share of the revenue. However, our picks and opinions are independent from USA TODAY's newsroom and any business incentives. Lots of people keep their Amazon Echo or other Alexa-enabled smart speaker in the kitchen, mostly because it's a central location in the home. However, this is actually a great place to use Alexa, as she has a host of under-appreciated kitchen skills. Sure, you can set timers and convert measurements with Alexa, but she can do so much more than that--from finding you recipes to maintaining your grocery list and suggesting wine pairings.

A wave of obsession for all-things machine learning (ML) has washed over the technology and business communities -- and society more broadly -- in the last several years, and understandably so; machine learning-enabled products and services can present myriad benefits to an organization -- not least the ability to harness large swaths of data to make previously tedious tasks more easy and efficient. Having a solid foundation for real-world ML is a major determinant of success for new initiatives, and is an exciting area of research and engineering in its own right, but the implementation of ML can even be challenging for organizations with mature engineering strength, and it goes without saying that there can be pitfalls and misconceptions in attempts to make the jump between machine learning research and ML in production environments. A frequently overshadowed and often under-appreciated aspect of getting it right is the infrastructure that enables robust, well-managed research and serves customers in production applications. A key lever in setting the foundation for a successful ML program is building a culture and an atmosphere that allows you to trial these efforts at scale: accelerating the rate of scientific experimentation on the road to production and, ultimately, to business value. The cloud is an integral part of these efforts, and it can enable teams to develop and deploy well-governed, accurate ML models to high-volume production environments.

The actual impact of artificial intelligence (AI) on the world's economy and jobs will likely be somewhere between the utopian and dystopian futures that it is often discussed in terms of, according to a new report from the Economist Intelligence Unit. The report, commissioned by Google, examined how AI will impact certain industries in the US, the UK, Australia, Japan, and Asia as a whole. The findings are based on econometric modelling, desk research, and interviews with academic and industry experts. Firms developing and using machine learning need to better communicate among themselves as well as with the public and policymakers, the report stated. This means doing more to manage expectations around the impact of machine learning, acknowledging the potential risks and rewards, improving trust and transparency, and educating the public.

The real face of Saint Valentine may have finally been revealed thanks to new 3D scans of the Patron Saint of Love's skull. Digital reconstructions reveal the Italian Saint had a strong jawline and died a young man when he was killed on February 14, 269 AD. The new scans contradict previous reconstructions of the Patron Saint of Love, which have typically suggested he was an older man of affluence. The real face of Saint Valentine may have finally been revealed thanks to new 3D scans of the Patron Saint of Love's skull. Little is known of the clergyman, but it is believed he was executed after he secretly defied a Roman Emperor's strict ban on marriage, helping soldiers to pair off with their wives illegally.

The House Transportation Policy committee voted 8-0 Wednesday to substitute an amendment that replaced proposed rules for the vehicles with a study group. That group would report to the Legislature in the fall.

Former NASA astronaut Mike Massimino reacts to the historic test flight. SpaceX is also building a new drone ship for rocket landings at sea, he added. When SpaceX's Falcon Heavy blasted off last Tuesday (Feb. The Falcon Heavy's two side boosters landed successfully (and simultaneously) on twin pads at the nearby Cape Canaveral Air Force Station, but the center core crashed and burned. That core booster, which was expected to land offshore on SpaceX's drone ship "Of Course I Still Love You," crashed when two of three engines did not fire during a final landing burn, Musk told reporters after the launch.

What do healthcare and finance have in common? Probably not too much, except for the fact that they are being disrupted by technology. The process is highly encouraged: a record sum of $3.5 billion was invested in 188 digital health companies in the first half of 2017. Yet, the key to the meaningful industry transformation lies in the use of data science for healthcare. With about 1.2 billion clinical documents being produced in the United States annually, life scientists and doctors have a sea of data to base their research upon. Additionally, huge volumes of health-related information are made accessible through widespread adoption of wearable tech. This opens up new opportunities for better, more informed healthcare. Being able to collect, structure and process a high volume of data and further make sense of it, to gain a deeper understanding of the human body is the key objective for thousands of data scientists and machine learning experts all over the world. It also has the strongest potential to revolutionize healthcare, based on our industry expertise.

Messages sent by aliens from space could destroy life as we know it on Earth if we're not careful about how we read them, scientists have warned. A new paper explores how we might read and understand a message that came to us from space. And it finds that it would be impossible to know that a message was dangerous before we opened it. In fact, the messages are so dangerous that it would be safest to simply discard them without ever reading them, scientists have warned. From the International Space Station, Expedition 42 Flight Engineer Terry W. Virts took this photograph of the Gulf of Mexico and U.S. Gulf Coast at sunset This image of an area on the surface of Mars, approximately 1.5 by 3 kilometers in size, shows frosted gullies on a south-facing slope within a crater.