Month: October 2021

Categories
Machine Learning Uncategorized

These neural networks know what they’re doing | Massachusetts Institute of Technology

Wangkun Jia/Shutterstock

From detecting cats in images to piloting a self-driving car, neural networks can learn to handle a wide range of issues. But it’s still unclear if these sophisticated pattern-recognition algorithms comprehend the tasks they’re executing. For example, rather than learning to recognize lanes and focus on the road’s horizon, a neural network tasked with maintaining a self-driving car in its lane might learn to do so by studying the shrubs on the side of the road. MIT researchers have demonstrated that a certain sort of neural network may understand the true cause-and-effect structure of the navigation task it is being trained to execute. 

Source: https://news.mit.edu/2021/cause-effect-neural-networks-1014 

Categories
Information Warfare

Air Force testing how to do intelligence in disconnected environments | C4ISRNET

BlueBarronPhoto/Shutterstock

The Air Force is experimenting with ways to conduct intelligence, surveillance, and reconnaissance in detached and dispersed settings as it works to improve its information warfare organization and position. Unlike the previous 20 years, the US military expects high-tech opponents to put pressure on every area of its operations, including the flow of data and information. In fact, rather than helping others at an event, the ISR community was the primary training audience for a previous activity, according to Haugh. The ISR elements are also thinking about how they can work with the Air Force’s agile combat employment concept, which defines how to launch and recover platforms swiftly away from key operating bases. 

Source: https://www.c4isrnet.com/information-warfare/2021/09/20/air-force-testing-how-to-do-intelligence-in-disconnected-environments/ 

Categories
Gaming Information Warfare

Epic Games Store Will ‘Welcome’ NFT Games After Steam Bans Them | C4ISRNET

Miguel Lagoa/Shutterstock

Even while NFTs continue to be lambasted as a big, planet-burning fraud populated with grifters, Epic Titles and its CEO Tim Sweeney are open to allowing games that utilize blockchain technology and NFTs onto its PC store. This comes after Valve announced that all games that use NFTs and blockchain technology would be removed and banned. It also follows Epic’s CEO’s previous assertions that the company will avoid dealing with NFTs owing to scams. Epic Games informed The Verge that it is open to collaborating with developers that want to incorporate blockchain technology, NFTs, and cryptocurrencies in games that may be sold on the publisher’s digital storefront, despite some limits. 

Source: https://kotaku.com/epic-games-store-will-welcome-nft-games-after-steam-ban-1847878298 

Categories
Drones Gaming

Drone pilot jobs gaining traction | The Star

Dmitry Kalinovsky/Shutterstock

The continuous vaccination campaign has not only opened up new economic options but has also provided new career chances, particularly for drone pilots. The employment of drone pilots is currently in demand, according to Malaysian Drone Club deputy president Wan Hasyidan Wan Hussein, as businesses, particularly tourism, are actively promoting their products through aerial films and images. According to him, a drone pilot might earn between RM500 and RM2,000 for each assignment, with the potential for more depending on the quality of work and types of footage, as well as risk concerns. 

Source: https://www.thestar.com.my/metro/metro-news/2021/09/21/drone-pilot-jobs-gaining-traction 

Categories
Cyber Security

Credit card PINs can be guessed even when covering the ATM pad | Bleeping Computer

Juta/Shutterstock

Even if the target is hiding the pad with their hands, researchers have demonstrated that a special-purpose deep-learning algorithm can guess 4-digit card PINs 41% of the time. Because programming the algorithm for the specific dimensions and key spacing of the individual PIN pads is critical, the attack necessitates the creation of a copy of the target ATM. Using footage of users typing PINs on the ATM pad, the machine-learning model is then taught to recognize pad presses and assign precise probabilities to a set of possibilities. The researchers collected 5,800 films of 58 different persons from various demographics for the experiment, using 4-digit and 5-digit PINs.