Welcome to the January 12, 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."
|
|
|
Smart Guns Arriving in the U.S.
Reuters
Daniel Trotta
January 11, 2022
Smart guns are beginning to become available to U.S. consumers, with smart gun maker LodeStar Works unveiling a 9-millimeter smart handgun. Smart guns could prevent accidental shootings, reduce suicides, and render lost or stolen guns worthless, as they use technology to authenticate the user’s identity and will disable a gun if an unauthorized person tries to fire it. Early prototypes used either fingerprint unlocking or a radio frequency identification (RFID) system that allows firing only when a chip in the gun interacts with a chip worn by the user. LodeStar's latest model combines a fingerprint reader, a phone application-activated near-field communication chip, and a personal identification number pad. Smart firearms from SmartGunz are secured by RFID, while Colorado-based Biofire is developing a smart gun incorporating a fingerprint reader.
|
Economists Pin More Blame on Tech for Rising Inequality
The New York Times
Steve Lohr
January 11, 2022
Some economists blame escalating inequality on the automation of tasks formerly done by humans, in addition to excessive technology investment and supportive public policies. Stanford University’s Erik Brynjolfsson warns of technologists, business people, and policymakers falling into “the Turing trap,” the assumption that artificial intelligence (AI) can match human performance, which leads to AI systems that replace people rather than augmenting their performance. Massachusetts Institute of Technology's Daron Acemoglu and Boston University's Pascual Restrepo determined "so-so technologies" that replace workers without raising productivity underlie sluggish productivity growth. Acemoglu endorses directing technology development along a more “human-friendly" path that works for, and with, people.
*May Require Paid Registration
|
Computer Model Seeks to Explain Spread of Misinformation, Suggests Countermeasures
Tufts Now
Mike Silver
January 10, 2022
A computer model developed by Tufts University researchers can mimic the real-life spread of misinformation, in order to expose its mechanisms and help in the development of countermeasures. The researchers based the model on the concept that pre-existing beliefs influence people's acceptance of new information; it assigns a numerical "belief" to each individual in an artificial social network scored from 0 to 6, with 6 representing strong acceptance. Tufts' Lenore Cowen suggested incorporating the realization that factual information alone may be insufficient to sway public mindsets into such models "may teach us how to bring the public conversation back to facts and evidence."
|
Robotic Arms Use ML to Reach Deeper into Distribution
The Wall Street Journal
Jennifer Smith
January 10, 2022
Robots increasingly are being used in warehouses to sort, pack, and prepare orders for delivery as logistics operators faced with labor shortages turn to automation to meet high demand. Advances in computer vision and software have allowed warehouse robots to take on more tasks previously handled by human workers. Puma North American Inc. is using robotic arms from Nimble Robotics Inc. to prepare clothing and shoe orders at a California distribution center, with plans to implement robots at another facility in Indiana. Puma's Helmut Leibbrandt said the robots can work two consecutive shifts and perform with about 99% accuracy, on par with human workers. Hasan Dandashly at logistics and manufacturing automation provider Dematic Corp. said it makes the greatest financial sense to use robots to pick orders in 24/7 operations with a limited number of products. Said Dandashly, "I don't think we are on the verge of not having human pickers anytime soon."
*May Require Paid Registration
|
Device Sniffs Out 'Smell-Fingerprints' of Pestered Plants
Agricultural Research Service
January 11, 2022
An electronic nose (E-Nose) developed by researchers at Ohio State University, the University of Tennessee-Knoxville, and the U.S. Department of Agriculture's Agricultural Research Service (ARS) can detect whitefly infestations of tomato plants. The prototype device sniffs out volatile organic compounds (VOCs) released by tomato plants in response to whitefly attacks. ARS' Heping Zhu said the E-Nose emulates mammalian olfactory perception and odor recognition; a nerve-like circuitry board translates VOC samples into digital signals, which are sent to an algorithm programmed to recognize the "smell-fingerprints" of VOCs emitted by tomato plants when attacked. The researchers hope the E-Nose could help greenhouse growers better time the use of insecticides and other infestation countermeasures.
|
Outsider Wins DARPA Challenge to Predict Where Floats Drift at Sea
New Scientist
David Hambling
January 10, 2022
A satellite engineer without oceanographic or meteorological expertise won the U.S. Defense Advanced Research Projects Agency (DARPA) Forecasting Floats in Turbulence Challenge, outperforming 31 other teams in predicting where floats will drift on the open sea. California-based Chris Wasson forecast the locations of 90 devices drifting in the Atlantic over 10 days, based on the previous 20 days of movement and meteorological information on currents, wind, and waves. Wasson simulated the effect of wind and surface currents on each float; for the first 20 days, he compared forecasts with actual positions to refine the model, using machine learning and mathematical modeling. Said Wasson, “Machine-learning approaches may suggest non-obvious solutions to problems and analytical methods can help to validate and explain those results."
|
AI Tool Could Help Diagnose Heart Failure
Imperial College London (U.K.)
Ellyw Evans
January 7, 2022
Combining a smart stethoscope with an artificial intelligence algorithm for early point-of-care heart failure diagnosis could improve patient outcomes at less cost, according to researchers at the U.K.'s Imperial College London (ICL). The algorithm, in conjunction with a stethoscope that records electrocardiograms and heart sounds, could determine a heart is exhibiting weak pumping action within 15 seconds, yielding 91% sensitivity and 80% specificity versus routine diagnostic tests. ICL's Patrik Bachtiger said, "This super-human capability to screen patients at any point of care, including the general practice surgery, can overcome the unacceptable reality that 80% of patients with heart failure are currently diagnosed through an emergency hospital admission."
|
Pet Tech at CES Treats Dogs, Cats Like Complex Beings
TechCrunch
Devin Coldewey; Haje Jan Kamps
January 5, 2022
This year's Consumer Electronics Show (CES) showcased gadgets designed to treat pets as complex, intelligent animals. For example, the Catlog collar and weight-sensing platform from Japanese developer Rabo helps owners monitor their cat's daily movements. The device perceives and correlates movement and vibrations with cat-like behavior like sleeping, sitting, cleaning itself, eating and drinking, and moving around; it sends this data to owners via an Internet of Things (IoT) hub. Meanwhile, French consumer electronics firm Invoxia offers a smart collar for dogs that tracks their location, heart rate, and breathing. FluentPet, meanwhile, makes kits for training dogs to express themselves in human language by pushing buttons on tiles that produce words, like "outside," "food, or "love you."
|
Nanowire Transistor with Integrated Memory to Enable Future Supercomputers
Lund University (Sweden)
January 10, 2022
Researchers at Sweden's Lund University have integrated a Resistive Random Access Memory cell with a vertical transistor selector at the nanoscale, resulting in faster calculations because they occur in the memory circuit itself. Lund's Lars-Erik Wernersson said, "Our version is a nanowire with a transistor at the bottom, and a very small memory element located further up on the same wire. This makes it into a compact integrated function where the transistor controls the memory element. The idea has been around before, but it has proven difficult to achieve performance. Now, however, we have shown that this can be achieved and that it works surprisingly well."
|
Quantum Advantage Next Goal in Race for New Computer Age
Financial Times
Richard Waters
January 4, 2022
Experts envision quantum advantage, the measured success of processing a real-world problem faster on a quantum computer than on a classic computer, as the next step in the race toward practical quantum computing applications. Matt Johnson at quantum software company QC Ware said, "The challenge is to be among the first to help enterprise customers get quantum speed-up." Technology developers suggest noisy intermediate-scale quantum systems could be tapped to make small advances toward real-world quantum apps. IBM in November released its first system using 127 quantum bits (qubits), and verified plans to exceed 1,000 qubits within two years. IBM CEO Chad Rigetti anticipates systemx featuring 4,000 qubits will be available in 2026, which "should carry us through the milestones of narrow and then broad quantum advantage."
*May Require Paid Registration
|
Scientists Use Summit Supercomputer, Deep Learning to Predict Protein Functions at Genome Scale
Oak Ridge National Laboratory
January 10, 2022
Scientists at the U.S. Department of Energy's Oak Ridge National Laboratory (ORNL), Google's DeepMind, and the Georgia Institute of Technology (Georgia Tech) are inferring genomic-scale protein functions using supercomputing and deep learning tools. The researchers used ORNL's Summit supercomputer, the fastest system in the U.S. and second-fastest globally, to model the full proteomes for four microbes, two of which generate valuable materials for manufacturing plastics, while the other two can break down and transform metals. A key computational tool developed at Georgia Tech, the Sequence Alignments from deep Learning of Structural Alignments, can compare genetic sequences by implicitly understanding protein structure, even when they have only 10% in common.
|
Algorithm Helps Robots Avoid Obstacles in Their Path
University of South Australia
January 10, 2022
An algorithm developed by researchers at the University of South Australia (UniSA) aims to help robots avoid humans and other obstacles in their path while taking the fastest, safest route to their destination. The researchers based their model on the best elements of existing algorithms and used it to create a TurtleBot able to avoid collisions by adjusting its speed and direction. They performed simulations in nine different scenarios and found their model outperformed the online collision avoidance algorithms Dynamic Window Approach and Artificial Potential Field. Said UniSA's Habib Habibullah, "Our proposed method sometimes took a longer path, but it was faster and safer, avoiding all collisions."
|
Tackling Hard Computational Problems
MIT News
Steve Nadis
January 10, 2022
The Massachusetts Institute of Technology's David Gamarnik and colleagues have developed the overlap gap property (OGP) tool to analyze difficult computational problems that involve randomness. "We discovered that all known problems of a random nature that are algorithmically hard have a version of this property," Gamarnik said. "This provides a more precise measure of algorithmic hardness." Scientists can evaluate the challenge of creating fast algorithms to solve particular problems with the OGP, and Gamarnik said the tool has already shown that stable algorithms, including quantum approximation optimization algorithms, cannot handle such problems.
|
|
