Once users have shared their data online, it is generally difficult for them to revoke access and ask for the data to be deleted. Machine learning (ML) exacerbates this problem because any model trained with said data may have memorized it, putting users at risk of a successful privacy attack exposing their information. Yet, having models unlearn is notoriously difficult. After a data point is removed from a training set, one often resorts to entirely retraining downstream models from scratch. We introduce SISA training, a framework that decreases the number of model parameters affected by an unlearning request and caches intermediate outputs of the training algorithm to limit the number of model updates that need to be computed to have these parameters unlearn. This framework reduces the computational overhead associated with unlearning, even in the worst-case setting where unlearning requests are made uniformly across the training set. In some cases, we may have a prior on the distribution of unlearning requests that will be issued by users. We may take this prior into account to partition and order data accordingly and further decrease overhead from unlearning. Our evaluation spans two datasets from different application domains, with corresponding motivations for unlearning. Under no distributional assumptions, we observe that SISA training improves unlearning for the Purchase dataset by 3.13x, and 1.658x for the SVHN dataset, over retraining from scratch. We also validate how knowledge of the unlearning distribution provides further improvements in retraining time by simulating a scenario where we model unlearning requests that come from users of a commercial product that is available in countries with varying sensitivity to privacy. Our work contributes to practical data governance in machine learning.

"It's fun," says research scientist Janelle Shane of her perpetual learning curve at Boulder Nonlinear Systems, a custom light-control manufacturing company. "This was my first job after my PhD. I knew I wanted to go into industry, and this merges post-doc-style research with business." With her colleagues, Shane works on projects that encompass a multitude of optics-related technologies, from nonmechanical beamsteering for planetary landers and self-driving cars, to ultrafast microscopy and spatial light modulators for neuroscientists. "We're driven by cutting-edge science and pushed to build something new," she says.

Facial recognition technology has advanced swiftly in the last five years. As University of Texas at Dallas researchers try to determine how computers have gotten as good as people at the task, they are also shedding light on how the human brain sorts information. UT Dallas scientists have analyzed the performance of the latest echelon of facial recognition algorithms, revealing the surprising way these programs -- which are based on machine learning -- work. Their study, published online Nov. 12 in Nature Machine Intelligence, shows that these sophisticated computer programs -- called deep convolutional neural networks (DCNNs) -- figured out how to identify faces differently than the researchers expected. "For the last 30 years, people have presumed that computer-based visual systems get rid of all the image-specific information -- angle, lighting, expression and so on," said Dr. Alice O'Toole, senior author of the study and the Aage and Margareta Møller Professor in the School of Behavioral and Brain Sciences.

By now, most of us have shared a few chuckles over AI-generated deepfake videos, like those in which the face of comedian and impressionist Bill Hader gradually takes on the likenesses of Tom Cruise, Seth Rogen, and Arnold Schwarzenegger as he imitates the celebrities. We've seen actor Ryan Reynolds' mug superimposed over Gene Wilder's in the 1971 classic film "Willy Wonka & the Chocolate Factory." We've even marveled over businessman Elon Musk being turned into a baby. It all can be quite humorous, but not everyone is laughing. Tech companies, researchers, and politicians alike are growing concerned that the increasing sophistication of the artificial intelligence and machine learning technology powering deepfakes will outpace our ability to discern between genuine and doctored imagery.

Almost every presentation began apologetically with the refrain, "In a 5G world" practically challenging the industry's rollout goals. At one point Brigitte Daniel-Corbin, IoT Strategist with Wilco Electronic Systems, sensed the need to reassure the audience by exclaiming, 'its not a matter of if, but when 5G will happen!' Frontier Tech pundits too often prematurely predict hyperbolic adoption cycles, falling into the trap of most soothsaying visions. The IoTC Summit's ability to pull back the curtain left its audience empowered with a sober roadmap forward that will ultimately drive greater innovation and profit. The industry frustration is understandable as China announced earlier this month that 5G is now commercially available in 50 cities, including: Beijing, Shanghai and Shenzhen.

Fox News Flash top headlines for Dec. 7 are here. Check out what's clicking on Foxnews.com The U.S. military believes the unarmed drone that went missing over the Libyan capital last month was actually shot down by Russian air defenses. The U.S. Africa Command is demanding the return of the aircraft's wreckage, which had been part of an operation conducted in Libya to assess the area's security and monitor for violent extremist activity. The command didn't give a reason for the drone loss after the Nov. 21 incident, but they had been investigating, Reuters reported.

Gartner predicts $137.4B will be spent on Information Security and Risk Management in 2019, increasing to $175.5B in 2023, reaching a CAGR of 9.1%. Cloud Security, Data Security, and Infrastructure Protection are the fastest-growing areas of security spending through 2023. Spending on AI-based cybersecurity systems and services reached $7.1B in 2018 and is predicted to reach $30.9B in 2025, attaining a CAGR of 23.4% in the forecast period according to Zion Market Research. Traditional approaches to securing endpoints based on the hardware characteristics of a given device aren't stopping breach attempts today. Bad actors are using AI and Machine Learning to launch sophisticated attacks to shorten the time it takes to compromise an endpoint and successfully breach systems.

Fortinet has unveiled predictions from the FortiGuard Labs team about the threat landscape for 2020 and beyond. These predictions reveal methods that Fortinet anticipates cybercriminals will employ in the near future, along with important strategies that will help organizations protect against these oncoming attacks. Changing the Trajectory of Cyberattacks Cyberattack methodologies have become more sophisticated in recent years magnifying their effectiveness and speed. This trend looks likely to continue unless more organizations make a shift as to how they think about their security strategies. With the volume, velocity, and sophistication of today's global threat landscape, organizations must be able to respond in real time at machine speed to effectively counter aggressive attacks.

Robotics is the field of engineering that is focused on developing machines – usually by combining knowledge from both the field of mechanics and electronics – that are able to either fully or partly take over tasks that would normally be carried out by humans. As technology advanced, components such as processors, electric motors and different kind of sensors have become increasingly more compact and precise, stimulating the use of robotics in other fields than only that of manufacturing. Robotics, fueled by artificial Intelligence (AI) is rapidly changing the way we work and interact with technology. Technological advances have drastically changed the capabilities of modern robotics, allowing them to be operational and effective in a wide range of industries. New technologies – such as artificial intelligence, machine learning, and advanced computer vision – form the basis of the further development of robotics, allowing them to be used in processes which were, up until today, deemed to only be executable by humans.

Robots have replaced thousands of routine jobs on Wall Street. That's the contention of Marcos Lopez de Prado, a Cornell University professor and the former head of machine learning at AQR Capital Management LLC, who testified in Washington on Friday about the impact of artificial intelligence on capital markets and jobs. The use of algorithms in electronic markets has automated the jobs of tens of thousands of execution traders worldwide, and it's also displaced people who model prices and risk or build investment portfolios, he said. "Financial machine learning creates a number of challenges for the 6.14 million people employed in the finance and insurance industry, many of whom will lose their jobs -- not necessarily because they are replaced by machines, but because they are not trained to work alongside algorithms," Lopez de Prado told the U.S. House Committee on Financial Services. During the almost two-hour hearing, lawmakers asked experts about racial and gender bias in AI, competition for highly skilled technology workers, and the challenges of regulating increasingly complex, data-driven financial markets.