Welcome to the May 5, 2021 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."
|
|
|
Ancient Australian 'Superhighways' Suggested by Massive Supercomputing Study
Science
David Malakoff
May 4, 2021
A multi-institutional team of researchers used supercomputers to plot the most likely migration routes of ancient humans across Australia. The team built the first detailed topographic map of the ancient Sahul landmass from satellite, aerial, and undersea mapping data, then calculated the optimal walking routes across this landscape via least-cost path analysis. Devin White at the U.S. Department of Energy's Sandia National Laboratories said a supercomputer operated by the U.S. government ran the simulations over weeks, which yielded a network of "optimal superhighways" featuring the most appealing combinations of easy walking, water, and landmarks. The University of Montana, Missoula's Kyle Bocinsky said, "This is a really compelling illustration of the power of using these [simulation] techniques, at a huge, continental scale, to understand how people navigate landscapes. It's impressive, extreme computing."
|
Speeding New COVID Treatments with Computational Tool
University of New Mexico
Michael Haederle
May 3, 2021
Scientists at the University of New Mexico (UNM) and the University of Texas at El Paso have developed a computational tool to help drug researchers quickly identify anti-COVID molecules before the virus invades human cells or disable it in the early stages of infection. The team unveiled REDIAL-2020, an open source suite of computational models that can help to rapidly screen small molecules for potential COVID-fighting traits. REDIAL-2020 is based on machine learning (ML) algorithms that quickly process massive volumes of data and tease out patterns that might be missed by human researchers. The team validated the ML forecasts by comparing datasets from the National Center for Advancing Translational Sciences to the known effects of approved drugs in UNM's DrugCentral database.
|
Algorithms Improve How We Protect Our Data
Daegu Gyeongbuk Institute of Science and Technology (South Korea)
May 3, 2021
Researchers at South Korea's Daegu Gyeongbuk Institute of Science and Technology (DGIST) have developed algorithms that estimate and validate encryption security with less computational complexity. The "Min-entropy" metric typically is used to estimate and validate a source's ability to generate random numbers used to encrypt data. An offline algorithm developed by the researchers estimates min-entropy based on an entire dataset; they also developed an online estimator that requires limited data samples and improves in accuracy as the data samples increase. Because the online estimator does not require storage for entire datasets, it is suitable for Internet of Things devices and other applications with memory, storage, and hardware constraints. DGIST's Yongjune Kim said, "Our evaluations showed that our algorithms can estimate min-entropy 500 times faster than the current standard algorithm while maintaining estimation accuracy."
|
ML Approach Brings Digital Photos Back to Life
Texas A&M Engineering News
Stephanie Jones
May 4, 2021
A machine learning (ML) technique developed by researchers at Texas A&M University (TAMU) enables users to produce novel views of a scene from a single photo. TAMU's Nima Kalantari said, "We can download and use any image on the Internet, even ones that are 100 years old, and essentially bring it back to life and look at it from different angles." Kalantari and graduate student Qinbo Li trained a deep learning network to generate novel views based on a single input image by showing it a set of over 2,000 images and corresponding novel-view images. Each input image was converted into a multiplane image, training the network to infer the location of objects in the scene.
|
Patch Issued to Tackle Critical Security Issues Present in Dell Driver Software Since 2009
ZDNet
Charlie Osborne
May 4, 2021
Computer vendor Dell has issued a patch to remedy five longstanding vulnerabilities in driver software discovered by a team at threat intelligence solutions provider SentinelLabs. Security researcher Kasif Dekel found the flaws by exploring the DBUtil BIOS driver found in Dell's desktop and laptop PCs, notebooks, and tablets. The focus of his investigation was the software's dbutil_2_3.sys module, which is installed and loaded on-demand by initiating the firmware update process, then unloaded after a system reboot. Two of the flaws identified were memory corruption issues in the driver, another two were security failures rooted in a lack of input validation, and the final issue found could be leveraged to trigger a denial of service. The SentinelLabs team said these vulnerabilities have been present since 2009, although there is no evidence of exploitation in the wild.
|
How Much Does It Itch?
Northwestern Now
April 30, 2021
A soft, wearable sensor developed by Northwestern University scientists can measure the itchiness suffered by children with atopic dermatitis (eczema), as well as by adults with liver disease, kidney disease, and certain cancers who suffer similar symptoms. The sensor quantifies itch by measuring scratching when placed on the hand, including finger-, wrist-, and elbow motion-related scratching. Incorporated into the device are machine learning algorithms that identify scratching without misidentifying similar motion-related movement. The sensor gauges both low-frequency motion and high-frequency vibrations from the hand to improve accuracy compared to wrist-watch tools. Said Northwestern’s Shuai Xu, “Patients with atopic dermatitis are 44% more likely to report suicidal thoughts as a result of the itch compared to controls. Thus, the ability to quantify their symptoms is really important to help new drugs get approved, but also support their day-to-day lives.”
|
Researchers Successfully Use 3D 'Bioprinting' to Create Nose Cartilage
Folio (University of Alberta, Canada)
Ross Neitz
May 4, 2021
A three-dimensional (3D) bioprinting technique developed by researchers at Canada's University of Alberta (U of A) can generate customized cartilage for use in restorative surgeries. The team employed a specially designed hydrogel that is combined with cells harvested from a patient, then printed in a specific configuration captured through 3D imaging. The material is cultured in a laboratory to become functional cartilage, which U of A's Adetola Adesida said can be ready for implantation within four weeks. Adesida said with this technology, a patient "can go on the operating table, have a small biopsy taken from their nose in about 30 minutes, and from there we can build different shapes of cartilage specifically for them. We can even bank the cells and use them later to build everything needed for the surgery."
|
Multinode Quantum Network is a Breakthrough for Quantum Internet
LiveScience
Ben Turner
May 3, 2021
Scientists at the Delft University of Technology (TU Delft) in the Netherlands have designed the world's first multinode quantum network, comprised of three entangled quantum bits (qubits) or nodes. The researchers used an intricate system of mirrors and laser light to produce and beam the entangled photons to the nodes correctly; mitigating noise and ensuring perfect laser synchronization were key challenges. TU Delft's Matteo Pompili said the multinode system "will allow us to connect quantum computers for more computing power, create unhackable networks, and connect atomic clocks and telescopes together with unprecedented levels of coordination."
|
Machine Vision System for Almond Grading, Safety
University of South Australia
May 3, 2021
An automated technique developed by researchers at the University of South Australia (UniSA) and industry partner SureNut can grade almond quality while detecting potential mycotoxin contamination in almond kernels. The researchers developed a machine equipped with two high-definition cameras, a hyperspectral camera, and purpose-developed artificial intelligence algorithms, which UniSA's Sang-Heon Lee said "can detect defects more quickly and more accurately than manual grading." Lee said the model correctly predicted moisture, free fatty acid content, and peroxide value, which are associated with rancidity, with accuracy rates of 95%, 93%, and 91%, respectively. It also correctly predicted aflatoxin B1 content with 94% accuracy.
|
Untangle Your Hair With Help From Robots
MIT Computer Science and Artificial Intelligence Laboratory
Rachel Gordon
May 3, 2021
Researchers at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Laboratory (CSAIL) and Harvard University's Soft Math Lab teamed up to develop a robotic arm setup that can comb tangled hair. "RoboWig" features a sensorized soft brush equipped with a camera to assess curliness, so the system can adapt to the degree of hair tangling it encounters. CSAIL's Josie Hughes said, "By developing a model of tangled fibers, we understand from a model-based perspective how hairs must be entangled: starting from the bottom and slowly working the way up to prevent 'jamming' of the fibers." Tests on wigs of various hair styles and hair types helped determine appropriate brushing lengths, taking into consideration the number of entanglements and pain levels.
|
Digital Decision-Aiding Tool to Personalize Choice of Test for Patients with Chest Pain
News-Medical Life Sciences
Emily Henderson
April 29, 2021
A new digital decision-aiding tool, ASSIST, developed by Yale University researchers uses machine learning to identify which of two imaging tests should be used on patients who may have coronary artery disease. The researchers applied machine learning techniques to data from two large clinical trials. They developed a novel strategy using data from 9,572 patients enrolled in the PROMISE trial that embedded local data experiments within the larger clinical trial. They also found that data on 2,135 patients in the SCOT-HEART trial who underwent functional first or anatomical-first testing showed the risk of adverse cardiac events was two-fold lower when the test performed and the one recommended by ASSIST were in agreement. Yale's Dr. Rohan Khera said, "A unique aspect of our approach is that we leverage both arms of a clinical trial, overcoming the limitation of real-world data, where decisions made by clinicians can introduce bias into algorithms."
|
Protocol Makes Bitcoin Transactions More Secure, Faster Than Lightning
TU Wein (Austria)
May 4, 2021
Researchers at Austria's Vienna University of Technology (TU Wein), Spain's IMDEA Software Institute, and Purdue University have developed an improved protocol for faster, more secure Bitcoin transactions. The researchers sought to improve on the "Lightning Network" of payment channels between blockchain users, which allows many transactions to be processed in a short amount of time. A simulation showed the new protocol results in a factor of four to 33 fewer failed transactions, compared with the Lightning Network. TU Wein's Lukas Aumayr said, "We can mathematically prove that our new protocol does not allow certain errors and problems in any situation."
|
Neural Nets Used to Rethink Material Design
Rice University News
April 30, 2021
A technique developed by researchers at Rice University and Lawrence Livermore National Laboratory uses machine learning to predict the evolution of microstructures in materials. The researchers demonstrated that neural networks can train themselves to predict a structure's growth in a particular environment. The researchers trained their neural networks using data from the traditional equation-based approach to predict microstructure changes and tested them on four microstructure types: plane-wave propagation, grain growth, spinodal decomposition, and dendritic crystal growth. The neural networks were 718 times faster for grain growth when powered by graphic processors compared to the prior algorithm, and 87 times faster when run on a standard central processor. Rice's Ming Tang said the new method can "make predictions even when we do not know everything about the material properties in a system," and will be useful in designing more efficient batteries.
|
|
