Subterms
Researchers have shown that an approach called federated learning is successful in the context of brain imaging, by being able to analyze magnetic resonance imaging (MRI) scans of brain tumor patients and distinguish healthy brain tissue from cancerous regions. More
New research has outlined three new approaches that digital innovators can take to reduce the risk of failure and seize competitive advantage in the industry. More
The question has been central to cryptography for thousands of years, and lies at the heart of efforts to secure private information on the internet. In a new paper, Cornell Tech researchers identified a problem that holds the key to whether all encryption can be broken — as well as a surprising connection to a mathematical concept that aims to define and measure randomness.
“Our result not only shows that cryptography has a natural ‘mother’ problem, it also shows a deep connection between two quite separate areas of mathematics and computer science — cryptography and algorithmic information theory,” said Rafael Pass, professor of computer science at Cornell Tech.
Pass is co-author of “On One-Way Functions and Kolmogorov Complexity,” which will be presented at the IEEE Symposium on Foundations of Computer Science, to be held Nov. 16-19 in Durham, North Carolina.
“The result,” he said, “is that a natural computational problem introduced in the 1960s in the Soviet Union characterizes the feasibility of basic cryptography — private-key encryption, digital signatures and authentication, for example.”
For millennia, cryptography was considered a cycle: Someone invented a code, the code was effective until someone eventually broke it, and the code became ineffective. In the 1970s, researchers seeking a better theory of cryptography introduced the concept of the one-way function — an easy task or problem in one direction that is impossible in the other.
advertisement
For example, it’s easy to light a match, but impossible to return a burning match to its unlit state without rearranging its atoms — an immensely difficult task.
“The idea was, if we have such a one-way function, maybe that’s a very good starting point for understanding cryptography,” Pass said. “Encrypting the message is very easy. And if you have the key, you can also decrypt it. But someone who doesn’t know the key should have to do the same thing as restoring a lit match.”
But researchers have not been able to prove the existence of a one-way function. The most well-known candidate — which is also the basis of the most commonly used encryption schemes on the internet — relies on integer factorization. It’s easy to multiply two random prime numbers — for instance, 23 and 47 — but significantly harder to find those two factors if only given their product, 1,081.
It is believed that no efficient factoring algorithm exists for large numbers, Pass said, though researchers may not have found the right algorithms yet.
“The central question we’re addressing is: Does it exist? Is there some natural problem that characterizes the existence of one-way functions?” he said. “If it does, that’s the mother of all problems, and if you have a way to solve that problem, you can break all purported one-way functions. And if you don’t know how to solve that problem, you can actually get secure cryptography.”
Meanwhile, mathematicians in the 1960s identified what’s known as Kolmogorov Complexity, which refers to quantifying the amount of randomness or pattern of a string of numbers. The Kolmogorov Complexity of a string of numbers is defined as the length of the shortest computer program that can generate the string; for some strings, such as 121212121212121212121212121212, there is a short program that generates it — alternate 1s and 2s. But for more complicated and apparently random strings of numbers, such as 37539017332840393452954329, there may not exist a program that is shorter than the length of the string itself.
advertisement
The problem has long interested mathematicians and computer scientists, including Juris Hartmanis, professor emeritus of computer science and engineering. Because the computer program attempting to generate the number could take millions or even billions of years, researchers in the Soviet Union in the 1960s, as well as Hartmanis and others in the 1980s, developed the time-bounded Kolmogorov Complexity — the length of the shortest program that can output a string of numbers in a certain amount of time.
In the paper, Pass and doctoral student Yanyi Liu showed that if computing time-bounded Kolmogorov Complexity is hard, then one-way functions exist.
Although their finding is theoretical, it has potential implications across cryptography, including internet security.
“If you can come up with an algorithm to solve the time-bounded Kolmogorov complexity problem, then you can break all crypto, all encryption schemes, all digital signatures,” Pass said. “However, if no efficient algorithm exists to solve this problem, you can get a one-way function, and therefore you can get secure encryption and digital signatures and so forth.”
The research was funded in part by the National Science Foundation and the Air Force Office of Scientific Research, and was based on research funded by the Intelligence Advanced Research Projects Activity in the Office of the Director of National Intelligence. More
The simplicity and elegance of origami, an ancient Japanese art form, has motivated researchers to explore its application in the world of materials.
New research from an interdisciplinary team, including Northwestern Engineering’s Horacio Espinosa and Sridhar Krishnaswamy and the Georgia Institute of Technology’s Glaucio Paulino, aims to advance the creation and understanding of such folded structures for applications ranging from soft robotics to medical devices to energy harvesters.
Inspired by origami, mechanical metamaterials — artificial structures with mechanical properties defined by their structure rather than their composition — have gained considerable attention because of their potential to yield deployable and highly tunable structures and materials.
What wasn’t known was which structures integrate shape recoverability, pronounced directional mechanical properties, and reversible auxeticity — meaning their lateral dimensions can increase and then decrease when progressively squeezed. Though some 3D origami structures have been produced through additive manufacturing, achieving the folding properties displayed in ideal paper origami remained a challenge.
Using nanoscale effects for an origami design, the team of researchers from the McCormick School of Engineering and Georgia Tech sought to answer that question. They produced small, 3D, origami-built metamaterials, successfully retaining the best properties without resorting to artifacts to enable folding.
“The created structures constitute the smallest fabricated origami architected metamaterials exhibiting an unprecedented combination of mechanical properties,” said Espinosa, the James and Nancy J. Farley Professor of Manufacturing and Entrepreneurship and professor of mechanical engineering and (by courtesy) biomedical engineering and civil and environmental engineering.
advertisement
“Our work demonstrated that rational design of metamaterials, with a large degree of shape recoverability and direction-dependent stiffness and deformation, is possible using origami designs, and that origami foldability enables a state where the material initially expands and subsequently contracts laterally (reversible auxeticity),” added Espinosa, who serves as director of Northwestern’s Theoretical and Applied Mechanics graduate program. “Such properties promise to influence a number of applications across a wide range of fields encompassing the nano-, micro-, and macro-scales, leveraging the intrinsic scalability of origami assemblies.”
“Guided by geometry, the scaling and miniaturization of the origami metamaterial are exciting in itself and by the unprecedented multifunctionality that it naturally enables,” said Paulino, the Raymond Allen Jones Chair at Georgia Tech’s School of Civil and Environmental Engineering.
“Only an interdisciplinary effort combining origami design, 3D laser printing with nanoscale resolution, and in situ electron microscopy mechanical testing could reveal the unprecedented combination of properties our work demonstrated and their potential impact on future applications,” added Paulino, who contributed to establishing the National Science Foundation Emerging Frontiers in Research and Innovation program named ODISSEI (Origami Design for Integration of Self-assembling Systems for Engineering Innovation).
“Just like nature has architected a wide range of structures using just a few material systems, origami allows us to engineer resilient structural components with distinct physical properties along different directions,” said Krishnaswamy, professor of mechanical engineering.
“We can envision origami-based soft microrobots that are stiff along some directions to carry payloads while maintaining other degrees of flexibility for motion. Origami-metamaterials that exploit reversible auxeticity and large deformation can lead to multifunctional applications ranging from deployable microsurgical instruments and medical devices, to energy steering and harvesting,” added Krishnaswamy, the director of Northwestern’s Center for Smart Structures and Materials.
The study presents new avenues to be explored long term, Espinosa said.
“There are a number of possibilities,” he said. “One is the fabrication of origami structures with ceramic and metallic materials, while preserving nanoscale dimensions, to exploit size effects in the mechanical response of the structures leading to superior energy dissipation per unit volume and mass. Another is the use of piezoelectric polymers, which can result in energy harvesters that can drive sensing modalities or power microsurgical tools.” More
Humans have a hard time identifying individual birds just by looking at the patterns on their plumage. An international study involving scientists form the CNRS, Université de Montpellier and the University of Porto in Portugal, among others, has shown how computers can learn to differentiate individual birds of a same species. The results are published on 27 July 2020 in Methods in Ecology and Evolution.
Differentiating between individuals of a same species is essential in the study of wild animals, their processes of adaptation and behaviour. Scientists from the CEFE research centre in Ecology and Evolutionary Ecology (CNRS/ Université de Montpellier/ Université Paul-Valéry-Montpellier/ IRD/ EPHE) and the Research Centre in Biodiversity and Genetic Resources (CIBIO) at Porto University have for the very first time identified individual birds with the help of artificial intelligence technology.
They have developed a technique that enables them to gather a large number of photographs, taken from various angles, of individual birds wearing electronic tags. These images were fed into computers which used deep learning technology to recognise the birds by analysing the photographs. The computers were able to distinguish individual birds according to the patterns on their plumage, something humans can’t do. The technology was able to identify specimens from populations of three different species: sociable weavers, great tits and zebra finches.
This new technique could not only result in a less invasive method of identification but also lead to new insights in ecology, for example, by opening ways of using AI to study animal behaviour in the wild.
Story Source:
Materials provided by CNRS. Note: Content may be edited for style and length. More
College students were more anxious and depressed during the initial outbreak of COVID-19 than they were during similar time frames in previous academic years, according to a Dartmouth study.
The research also found that sedentary behavior increased dramatically during the onset of the public health crisis in early March.
The study, published in the Journal of Medical Internet Research, used a mix of smartphone sensing and digital questionnaires from more than 200 students participating in a research program that is tracking mental health throughout their undergraduate years.
“COVID-19 had an immediate negative impact on the emotional well-being of the college students we studied,” said Jeremy Huckins, a lecturer on psychological and brain sciences at Dartmouth. “We observed a large-scale shift in mental health and behavior compared to the observed baseline established for this group over previous years.”
Self-reported symptoms of depression and anxiety within the student research group spiked noticeably at the onset of COVID-19. At the time, major policy changes related to COVID-19 were also being put in place, including the request that students leave campus and the switch to remote learning.
These changes coincided with the end of classes and final exams, already one of the most stressful times for students in any academic term.
advertisement
According to the study, anxiety and depression decreased slightly after the final exam period as students settled into shelter-in-place locations. This suggested some resilience in the face of COVID-19, but levels remained consistently higher than similar periods during previous academic terms.
Unlike previous terms studied, sedentary time increased dramatically during this year’s spring break period.
“This was an atypical time for these college students. While spring break is usually a period of decreased stress and increased physical activity, spring break 2020 was stressful and confining for the students participating in this study. We suspect that this was the case for a large number of college students across the country,” said Huckins.
The study used StudentLife, a sensing app developed at Dartmouth, to collect information from student volunteers. StudentLife passively collects behavioral information from user’s smartphones such as duration of phone usage, number of phone unlocks, sleep duration, and sedentary time.
Data on depression and anxiety were collected using weekly, self-reported assessments also administered through the StudentLife app.
advertisement
“This is the first time we have used sensor data from phones to give us unique behavioral insights into the reaction of students to the onset of the pandemic on a college campus,” said Andrew Campbell, the Albert Bradley 1915 Third Century Professor of computer science at Dartmouth and one of the lead researchers of the StudentLife study. “We plan to further analyze how these students adjusted both physically and mentally during remote learning that leads on from this study.”
In the research, the team also reported a connection between anxiety and COVID-19 news coverage. The link between depression and news reporting was apparent, but not as strong. As news coverage intensified, there was an increase in sedentary behavior and a longer duration of phone usage.
According to the study, the decrease in the number of locations visited was consistent with the social distancing and shelter-in-place policies implemented by local governments.
The study’s findings on the uptake of social distancing recommendations contrasts with other research of college students in which governmental social distancing policies were not followed. Findings in the current study are also contrary to media depictions of college-age students flouting social distancing recommendations during the spring break period.
“Many people wouldn’t expect college students to listen to social distancing orders, but these students did. We found that when social distancing was recommended by local governments, students were more sedentary and visited fewer locations on any given day,” said Huckins. “Clearly the impact of COVID-19 extends beyond the virus and its direct impacts. An unresolved question is if mental health and physical activity will continue to degrade over time, or if we will see a recovery, and how long that recovery will take.”
The research is part of a multiyear study focusing on the mental health of undergraduate students as they progress through their undergraduate careers. The complete study combines smartphone mobile sensing with functional neuroimaging.
“When we set out two years ago to follow 200 students across their college experiences, we could never have anticipated the inflection point in our data as a result of such a catastrophic event as the pandemic,” added Campbell.
Upon completion of the full study, researchers will be able to extend their findings on the disruption at the start of the COVID-19 pandemic to the long-term impact of remote learning and social isolation that the students are experiencing.
More information on the StudentLife research program can be found at: https://studentlife.cs.dartmouth.edu More
Repeated activity wears on soft robotic actuators, but these machine’s moving parts need to be reliable and easily fixed. Now a team of researchers has a biosynthetic polymer, patterned after squid ring teeth, that is self-healing and biodegradable, creating a material not only good for actuators, but also for hazmat suits and other applications where tiny holes could cause a danger.
“Current self-healing materials have shortcomings that limit their practical application, such as low healing strength and long healing times (hours),” the researcher report in today’s issue of Nature Materials.
The researchers produced high-strength synthetic proteins that mimic those found in nature. Like the creatures they are patterned on, the proteins can self-heal both minute and visible damage.
“Our goal is to create self-healing programmable materials with unprecedented control over their physical properties using synthetic biology,” said Melik Demirel, professor of engineering science and mechanics and holder of the Lloyd and Dorothy Foehr Huck Chair in Biomimetic Materials.
Robotic machines from industrial robotic arms and prosthetic legs have joints that move and require a soft material that will accommodate this movement. So do ventilators and personal protective equipment of various kinds. But, all materials under continual repetitive motion develop tiny tears and cracks and eventually break. Using a self-healing material, the initial tiny defects are repairable before catastrophic failure ensues.
Demirel’s team creates the self-healing polymer by using a series of DNA tandem repeats made up of amino acids produced by gene duplication. Tandem repeats are usually short series of molecules arranged to repeat themselves any number of times. The researchers manufacture the polymer in standard bacterial bioreactors.
advertisement
“We were able to reduce a typical 24-hour healing period to one second so our protein-based soft robots can now repair themselves immediately,” said Abdon Pena-Francelsch, lead author of the paper and a former doctoral student in Demirel’s lab. “In nature, self-healing takes a long time. In this sense, our technology outsmarts nature.”
The self-healing polymer heals with the application of water and heat, although Demirel said that it could also heal using light.
“If you cut this polymer in half, when it heals it gains back 100 percent of its strength,” said Demirel.
Metin Sitti, director, Physical Intelligence Department at the Max Planck Institute for Intelligent Systems, Stuttgart, Germany, and his team, were working with the polymer, creating holes and healing them. They then created soft actuators that, through use, cracked and then healed in real time — about one second.
“Self-repairing physically intelligent soft materials are essential for building robust and fault-tolerant soft robots and actuators in the near future,” said Sitti.
advertisement
By adjusting the number of tandem repeats, Demirel’s team created a soft polymer that healed rapidly and retained its original strength, but they also created a polymer that is 100% biodegradable and 100% recyclable into the same, original polymer.
“We want to minimize the use of petroleum-based polymers for many reasons,” said Demirel. “Sooner or later we will run out of petroleum and it is also polluting and causing global warming. We can’t compete with the really inexpensive plastics. The only way to compete is to supply something the petroleum based polymers can’t deliver and self-healing provides the performance needed.”
Demirel explained that while many petroleum-based polymers can be recycled, they are recycled into something different. For example, polyester t-shirts can be recycled into bottles, but not into polyester fibers again.
Just as the squid the polymer mimics biodegrades in the ocean, the biomimetic polymer will biodegrade. With the addition of an acid like vinegar, the polymer will also recycle into a powder that is again manufacturable into the same, soft, self-healing polymer.
“This research illuminates the landscape of material properties that become accessible by going beyond proteins that exist in nature using synthetic biology approaches,” said Stephanie McElhinny, biochemistry program manager, Army Research Office, an element of the U.S. Army Combat Capabilities Development Command’s Army Research Laboratory. “The rapid and high-strength self-healing of these synthetic proteins demonstrates the potential of this approach to deliver novel materials for future Army applications, such as personal protective equipment or flexible robots that could maneuver in confined spaces.” More
Researchers at Dartmouth, in collaboration with industry partners, have developed software techniques that make lighting in computer-generated images look more realistic. The research will be presented at the upcoming ACM SIGGRAPH conference, the premier venue for research in computer graphics.
The new techniques focus on “real time” graphics which need to maintain the illusion of interactivity as scenes change in response to user moves. These graphics can be used in applications such as video games, extended reality, and scientific visualization tools.
Both papers demonstrate how developers can create sophisticated lighting effects by adapting a popular rendering technique known as ray tracing.
“Over the last decade, ray tracing has dramatically increased the realism and visual richness of computer-generated images in movies where producing just a single frame can take hours,” said Wojciech Jarosz, an associate professor of computer science at Dartmouth who served as the senior researcher for both projects. “Our papers describe two very different approaches for bringing realistic ray-traced lighting to the constraints of real time graphics.”
The first project, developed with NVIDIA, envisions the possibilities for future games once developers incorporate NVIDIA’s hardware-accelerated RTX ray tracing platform. Recent games have started to use RTX for physically correct shadows and reflections, but quality and complexity of lighting is currently limited by the small number of rays that can be traced per frame.
The new technique, called reservoir-based spatiotemporal importance resampling (ReSTIR), creates realistic lighting and shadows from millions of artificial light sources. The ReSTIR approach dramatically increases the quality of rendering on a computer’s graphics card by reusing rays that were traced in neighboring pixels and in prior frames.
advertisement
The new technique can be integrated into the design of future games and works up to 65 times faster than previous rendering techniques.
“This technology is not just exciting for what it can bring to real-time applications like games, but also its impact in the movie industry and beyond,” said Benedikt Bitterli, a PhD student at Dartmouth who served as the first author of a research paper on the technique.
The second project, conducted in collaboration with Activision, describes how the video game publisher has incorporated increasingly realistic lighting effects into its games.
Traditionally, video games create lighting sequences in real time using what are called “baked” solutions: the complex ray-traced illumination is computed only once through a time-consuming process. The lighting created using this technique can be displayed easily during gameplay, but it is constrained to assuming a fixed configuration for a scene. As a result, the lighting cannot easily react to the movement of characters and cameras.
The research paper describes how Activision gradually evolved its “UberBake” system from the static approach to one which can depict subtle lighting changes in response to player interactions, such as turning lights on and off, or opening and closing doors.
advertisement
Since UberBake was developed over many years to work on current games, it needed to work on a variety of existing hardware, ranging from high-end PCs to previous-generation gaming consoles.
“Video games are used by millions of people around the world,” said Dario Seyb, a PhD student at Dartmouth who served as the research paper’s co-first author. “With so many people interacting with video games, this technology can have a huge impact.”
Dartmouth researchers on both projects are affiliated with the Dartmouth Visual Computing Lab.
“These industry collaborations have been fantastic. They allow our students to work on foundational academic research informed by practical problems in industry, allowing the work to have a more immediate, real-world impact,” said Jarosz.
The research papers will be published in ACM Transactions on Graphics and presented at SIGGRAPH 2020 taking place online during the summer.
Story Source:
Materials provided by Dartmouth College. Note: Content may be edited for style and length. More
