Welcome to the April 11, 2022, edition of ACM TechNews, providing timely information for IT professionals three times a week.
ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).
To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version.
The online version now has a button at the top labeled "Show Headlines."
|
|
|
Engineer Makes Footgear to Guide People with Vision Problems
Asahi Shimbun (Japan)
Akiko Suzuki
April 8, 2022
Japanese engineer Wataru Chino has designed footwear that incorporates a flexible motor-mounted device embedded to guide vision-impaired users. Wearers of the Ashirase walk support system are directed to destinations by a specialized smartphone application, which guides them via vibrations from the device in the shoes. Chino was involved in the development of driverless vehicles and other systems at automaker Honda Motor, but realized in the course of his work "that walking could also be dangerous." After learning that bicycles left on tactile paving could impede visually impaired pedestrians, he was struck with the concept of "embedding braille information in footwear."
|
Researchers Find Clues to Mechanisms for Coronavirus Infections
WSU Insider
Tina Hilding
April 8, 2022
Washington State University researchers have uncovered insights into cross-species coronavirus transmissibility that could be applied to prevent future pandemics, using a computational approach based on network science. The researchers assembled network maps showing how viral spike sequences relate in order to better distinguish between coronaviruses that can and cannot infect human cells. Concentrating on a small part of the spike protein used by some coronaviruses to bind to receptors revealed that the network map organized the viruses into human- and non-human-transmissible clusters. The researchers' laboratory results validated the network map's accuracy, and they hope this research will support improved vaccine development.
|
U.S. FBI Says It Disrupted Russian Hackers
Reuters
Sarah N. Lynch
April 6, 2022
U.S. officials said the Federal Bureau of Investigation (FBI) seized control of thousands of routers and firewall appliances from Russian hackers by appropriating the infrastructure used to communicate with the devices. An unsealed redacted affidavit said the operation attempted to prevent the hackers from networking the devices into a botnet with which they could assail other servers with rogue traffic. Said U.S. Attorney General Merrick Garland, "Fortunately, we were able to disrupt this botnet before it could be used.” The botnet was governed by Cyclops Blink malware, which U.S. and U.K. cyberdefense agencies had publicly attributed to Sandworm, a group associated with Russian military intelligence. FBI Director Chris Wray said, "We removed malware from devices used by thousands of mostly small businesses for network security all over the world. We shut the door the Russians had used to get into them."
|
Wireless Tech Lets Implants Transmit Data Through Ions in Tissue
New Atlas
Michael Irving
April 6, 2022
Columbia University scientists have devised a method of enabling medical implants to transmit data through ions in human tissue. A pair of implanted electrodes encode data from a device in alternating electric pulses, then store the energy in the ions within the tissue; a second electrode pair can be laid atop the skin to receive the stored energy and decode the data. The researchers said this technique can transmit data through tissue as deep as 10 centimeters (3.9 inches), with less signal degradation than radio-frequency communication. They also recorded transmission rates of up to 60 MHz, outperforming other methods. Tests using rats with neural interface implants showed the ionic communication system could transmit data to external receivers with sufficient precision to distinguish signals from individual neurons.
|
Smarter 3D Printing Makes Better Parts Faster
University of Michigan News
April 8, 2022
Software developed by University of Michigan (U-M) researchers could accelerate three-dimensional (3D) printing of intricate metal and plastic components by reducing heat buildup in laser powder bed fusion printers. The SmartScan software gauges how heat flows within a given part, and maps an optimized scan sequence to restrict heat accumulation in any given area. It analyzes the part's contours and the thermal properties of the material being used. U-M's Chinedum Okwudire calls SmartScan the first tool to use a thermal model to optimally direct the laser for more even heat distribution. The researchers found SmartScan imprinted patterns on stainless steel plates with less warping and more uniform heat distribution than traditional printing patterns.
|
Does This AI Think Like a Human?
MIT News
Adam Zewe
April 6, 2022
Massachusetts Institute of Technology (MIT) and IBM Research scientists have developed the Shared Interest method for rapidly analyzing a machine learning model's behavior by evaluating its individual explanations. The technique uses saliency methods to highlight how the model made specific decisions, comparing them to ground-truth data. Shared Interest then applies quantifiable metrics that compare the model's reasoning to that of a human by measuring the alignment between its decisions and the ground truth, then classifying those decisions into eight categories. The method can be used for image and text classification. MIT's Angie Boggust warned that the technique is only as good as the saliency methods on which it is based; if those techniques are biased or contain inaccuracies, the technique will inherit those limitations.
|
Team Tracks Overlap Between Sharks, Game Fish
UMass Amherst News
April 5, 2022
An international team of researchers led by the University of Massachusetts Amherst (UMass Amherst) has compiled a vast dataset on the positions, migrations, and interactions of sharks and game fish. The researchers deployed acoustic receivers and tagged 257 fish, including 73 sharks, with transmitters along the coast of the Florida Keys. They processed raw data through a machine learning algorithm to simulate the interplay of environmental factors. UMass Amherst's Grace Casselberry said, "Combining acoustic telemetry and machine learning helped us to answer a host of questions about predators and prey." The results have immediate value for anglers, who have been reporting growing contacts with sharks for years. Casselberry believes fisheries managers can use this information to "tailor their management strategies to best protect the interests of sharks, game fish, and anglers."
|
Better, Faster, Energy Efficient Predictions
Harvard University John A. Paulson School of Engineering and Applied Sciences
Leah Burrows
April 7, 2022
Researchers at Harvard University's John A. Paulson School of Engineering and Applied Sciences (SEAS) have developed "intelligent alloys" that combine the power of computation with artificial intelligence to create models that complement predictive evolutionary simulations. SEAS' Petros Koumoutsakos and Jane Bae computed turbulent flows by combining reinforcement learning and numerical methods, using machine learning (ML) agents that interact with mathematical equations. "We take an equation and play a game where the agent is learning to complete the parts of the equations that we cannot resolve," Bae said. "The agents add information from the observations the computations can resolve and then they improve what the computation has done."
|
Keywords Can Hack the Hiring Process
University of Texas at Arlington
Herb Booth
April 7, 2022
The University of Texas at Arlington (UTA)'s Shirin Nilizadeh found that an algorithm that uses job-specific keywords can help applicants improve their position by at least 16 spots on average in a pool of 100 applicants. "We found out that you can tailor your resume for a specific job by using specific keywords that could get you pushed toward the top," she explained. Text-embedding algorithms pair words and sentences in resumes with the job description to produce similarity scores on which resumes are ranked. Nilizadeh found that while adding more keywords improves the ranking, adding too many might not. UTA's Hong Jiang suggested Nilizadeh's work "might be a tool prospective employees and employers could use in the job search process."
|
Robot Learns to Open Doors by Splitting Task into Three Easy Steps
New Scientist
Matthew Sparkes
April 6, 2022
Researchers at Japan's Waseda University have developed a method that reduces the time and effort necessary to train a robot to open doors by dividing the deep learning model that controls the robot into modules. The task was split into three steps—approaching, opening, and passing through the door—with a module for each task devoted to doors that need to be pulled open and another for doors that are pushed open. The robot received six hours of training for each of the six modules and watched a human perform the task 108 times, after which it was able to complete the task 96% of the time.
*May Require Paid Registration
|
Hackers Target Bridges Between Blockchains for Crypto Heists
The Wall Street Journal
David Uberti
April 5, 2022
Hackers pulled off a $540-million cryptocurrency heist by exploiting Ronin Network software that lets players of the online game "Axie Infinity" transfer digital assets across blockchains. Ronin developer Sky Mavis said the hackers used a social engineering exploit to acquire the five keys needed to access Axie Infinity's underlying bridge. Blockchain analytics firm Elliptic estimates that decentralized financial systems saw at least $10.5 billion in losses last year due to crime, including stolen funds and falling prices in cryptocurrency offered by hacked systems. Blockchain experts said with cross-chain projects attracting capital and energy, security tools are under pressure to keep up.
*May Require Paid Registration
|
|
