Institute for Advanced Computer Studies
Institute for Advanced Computer Studies
Fermuller C.,Institute for Advanced Computer Studies |
Ji H.,National University of Singapore |
Kitaoka A.,Ritsumeikan University
Vision Research | Year: 2010
A new class of patterns, composed of repeating patches of asymmetric intensity profile, elicit strong perception of illusory motion. We propose that the main cause of this illusion is erroneous estimation of image motion induced by fixational eye movements. Image motion is estimated with spatial and temporal energy filters, which are symmetric in space, but asymmetric (causal) in time. That is, only the past, but not the future, is used to estimate the temporal energy. It is shown that such filters mis-estimate the motion of locally asymmetric intensity signals at certain spatial frequencies. In an experiment the perception of the different illusory signals was quantitatively compared by nulling the illusory motion with opposing real motion, and was found to be predicted well by the model. © 2009 Elsevier Ltd. All rights reserved.
News Article | November 16, 2016
Cambridge, Massachusetts, November 16, 2016-- When the widely respected Whitehead Institute researcher Susan L. Lindquist died on October 27, she left behind a legacy of both soaring scientific accomplishment and deep personal commitment. In a career spanning four decades, she changed the way scientists viewed the role of cellular proteins in human health, evolution, and biomaterials. She was also a leader who--through the example of her career and through her personal engagement--fostered the careers of women determined to fulfill their potential as scientists. To honor Dr. Lindquist, Johnson & Johnson has endowed the Susan Lindquist Chair for Women in Science at Whitehead Institute, to be awarded to a distinguished female scientist who is advancing biomedical research. "The biomedical research enterprise has been immeasurably enriched by the hundreds of women and men who, directly or indirectly, were inspired by Sue's achievements, guided by her wisdom, and nurtured by her support," observes David C. Page, M.D., Director of Whitehead Institute and Professor of Biology at the Massachusetts Institute of Technology (MIT). "She played a particularly important role as an exemplar and an advocate for women in the sciences. And we are immensely pleased that--through this professorship--Johnson & Johnson is honoring her legacy and advancing her vision of scientific research as a gender-neutral endeavor." In addition to serving as a Member of Whitehead Institute and Professor of Biology at MIT, Lindquist was a member of the Johnson & Johnson Board of Directors since 2004, serving as Chairman of its Science, Technology and Sustainability Committee and a member of its Regulatory, Compliance, and Government Affairs Committee. "Sue was a prolific scientific pioneer who changed fundamental understanding of the biology of human health. As part of the Johnson & Johnson Board of Directors, she challenged us to use science and technology in new ways to help improve the health and lives of people all around the world," says Alex Gorsky, Chairman and Chief Executive Officer of Johnson & Johnson. "We are pleased to establish this Chair in Sue's name, recognizing a greatly respected and beloved scientist and a passionate advocate for women in science." "Susan was a force of nature and a force for good," reflects Cori Bargmann, President of Science for the Chan Zuckerberg Initiative. Legions of leading female scientists credit Lindquist with helping advance their careers. "Sue Lindquist understood that pioneering scientists need champions--especially if the pioneering innovator is a woman. This chair creates a new link in the ever-growing chain of women scientists connecting Sue with women before and after her--women who mentor and promote one another as we work toward our shared dream of a profession without gender bias," says Bonnie Bassler, Ph.D., Squibb Professor and Chair in Molecular Biology at Princeton University, Howard Hughes Medical Institute Investigator (HHMI), MacArthur Fellowship winner, and member of the National Science Board appointed by President Obama. "In my own case, Sue heard an early seminar of mine, knew that my post-doc mentor had just retired, and took it upon herself to become my advocate. This selfless act was the most meaningful and important influence in my career." Recent MacArthur Foundation Fellowship winner Dianne K. Newman, Gordon M. Binder/Amgen Professor of Biology and Geobiology at California Institute of Technology, recalls that she first heard Lindquist speak when Newman was in graduate school. "She was so passionate about her work, so eloquent, and exemplified the type of daring, open minded, rigorous scientist I hoped one day to become," Newman says. "Many years later, as a colleague at MIT, I saw an equally impressive side of Sue: her personal warmth and generosity. Sue demonstrated that one could be a first-class scientist and a first-class human being. She was a phenomenal role model in all ways, to young men and women alike." The newly established chair will be held by a distinguished female member of the Whitehead Institute faculty, whose title would be Susan Lindquist Professor for Women in Science. "The title's phrasing--'for women in science'--was specifically chosen by Sue and endorsed by Whitehead Institute leadership," Page explains. "It is deliberately open-ended and consciously provocative. We want the incumbents to have great flexibility in how they pursue Sue's legacy of stellar science and courageous leadership. We hope it spurs discussion in the broader science community, and are comfortable if it sparks controversy." Brit D'Arbeloff, one of the few women to earn a graduate degree at MIT in the early 1960s, and her late husband, former Whitehead Institute Board Chair Alex D'Arbeloff, were close to Lindquist. "Sue was the reason Alex and I were involved with Whitehead," she recalls. "Sue was both an unbelievable researcher and an amazing, warm, open and interesting person. I miss her, but am thankful for having known her and seen all she was able to accomplish." Reflecting on Lindquist's stellar science, Cori Bargmann, says, "The beauty of her own scientific trajectory shows how depth and focus lead to advances: first asking how chaperones help proteins fold, then learning how misfolded yeast proteins propagate across generations, then using that knowledge to probe protein misfolding in human neurodegenerative diseases. And her creativity! Rutherford and Lindquist, 1998 ("Hsp90 as a capacitor for morphological evolution") will always be one of my favorite papers." "Susan Lindquist was a wonderful scientist who cared about those for whom her research was being done as much as the science itself," says Rita Colwell, Professor in the Institute for Advanced Computer Studies at University of Maryland, who was the first woman to serve as Director of the National Science Foundation. "Sue was noble in spirit and in action; that was most evident in her commitment to STEM education and in advocating for women in science. She left us too soon but we will long remember her contributions to the betterment of humanity through her scientific research, teaching, and mentoring." Family members have suggested that donations in Susan Lindquist's honor be directed to the Whitehead Institute Fund to Encourage Women in Science. Programmatic efforts funded by this fund will complement the new Lindquist Professor's mentoring and public outreach efforts, and may include K-12 programs, a lecture series, or a symposium for women in science (https:/ ).
Samet H.,Institute for Advanced Computer Studies |
Sankaranarayanan J.,NEC Labs |
Sankaranarayanan J.,University of Maryland University College |
Lieberman M.D.,Johns Hopkins University |
And 8 more authors.
Communications of the ACM | Year: 2014
The advantage of the map-query interface is that a map, coupled with the ability to vary the zoom level at which it is viewed, provides inherent granularity to a search process that facilitates approximate search. This capability distinguishes it from prevalent keyword-based conventional search methods that provide a limited facility for approximate searches that are realized primarily by permitting a match through a subset of the keywords. Being able to use spatial synonyms is important, as it enables users to search for data when they are not exactly sure what they seek or what the answer to their query should be. NewsStand's goal is to offer an alternative to the news-reading process and, more important, experience. Users can also constrain the spatial region and news sources; they need not be the same. This is a useful feature, as it enables users to see how one part of the world views events in another part of the world.
News Article | November 27, 2016
The future looks grim for the rhinoceros. Just 29,000 rhinos now exist in the wild, down from half a million at the beginning of the 20th century. Look closer, and the numbers become even more disheartening. The western black rhino was officially declared extinct in 2011. Three of the five remaining species aren't far behind: Just 58 to 61 Javan rhinos and fewer than 100 Sumatran rhinos now live in the wild. Armed guards in Kenya constantly protect the last three northern white rhinos on the planet. The animals' most lethal predators are human, attracted to the big business of rhino horn. And it is big business. By most estimates, rhino horn can fetch as much as $60,000 per kilo, or about $27,000 a pound. That makes it more valuable by weight than gold or cocaine. The chief markets are in Asia, where traditional medicine has used the horn for centuries to treat everything from fevers and convulsions to rheumatism and food poisoning. But demand for the horn has skyrocketed in just the past 10 years, primarily in Vietnam. Now the world's biggest consumer of rhino horn, the country's swelling well-to-do class prizes it as a status symbol of wealth and power, and a miracle cure for cancer. As a result, poaching has surged to "unprecedented levels," Save the Rhino International (SRI) says on its website. In South Africa alone, poachers slaughtered almost 3,400 rhinos over three years -- a rate of one animal every eight hours -- according to SRI. If the pace continues, rhinos could be wiped from the wild within the next decade. That could have far-reaching consequences because the rhinoceros is what scientists call an umbrella species. Protect it, and you protect the other species sharing its habitat. Rhinoceros survival matters -- and conservationists are turning to science and technology to save it. Several companies, including Ceratotech, Rhinoceros Horn LLC and Pembient, believe they have the solution for stopping illegal rhino horn traffic: Give consumers lab-grown alternatives at a fraction of the price, crowding the real thing out of the market. Pembient is the most prominent company in this space. Its approach relies on 3D bioprinting -- basically adding rhino DNA to synthetic keratin, then creating a sort of keratin ink that can run through a 3D printer. The company says its bioprinted material is genetically identical to real horn. Co-founder and CEO Matthew Markus thinks Pembient can do more good by flooding the market with lab-grown horn than traditional conservation efforts can. "When you show up in a country and say you can't use tiger bone, rhinoceros horn, pangolin scales and so on, that's a tough sell," says Markus. "We like to say Pembient is founded on the belief that animals are precious and traditions are important. I see value in both, while it seems most conservationists don't." Conservationists don't just see things differently. Over the past two years, more than a dozen organizations have written articles, published position papers or filed petitions against the sale of synthetic horn. This past February, WildAid and the Center for Biological Diversity petitioned the US Department of the Interior to ban the import, export and sale of bioengineered horn. Nearly all say fakes will just make things worse -- stimulating demand for real horn and reinforcing the myth that it can cure cancer. "What Pembient does is validate the untruth that there's any medical value to rhino horn," says CeCe Sieffert, deputy director of the International Rhino Foundation (IRF). "Having something created as a supplement or replacement says there is value." Thomas Snitch, previously executive officer of the UN Wildlife Enforcement Monitoring System and professor at the University of Maryland's Institute for Advanced Computer Studies, believes black marketers would just sell synthetic horn as the real thing. "The criminal syndicates would like to kill every rhino on the planet and control every rhino horn left in existence," he says. "Then a horn will have an infinite value. They will buy up the Pembient horn and sell it for tens of millions." This year, researchers with San Diego Zoo Global and the Leibniz Institute for Zoo and Wildlife Research, in Berlin, revealed they're working on an absolute last-ditch effort -- building rhinoceroses from scratch using stem cell technology. Here's how it would work. A team from the San Diego Zoo would induce stem cells from the three remaining northern white rhinos -- which are too old to breed -- into sperm and egg cells. The team will also use frozen sperm and other cells taken from 10 other northern white rhinos before they died. The scientists will then use IVF to fertilize the egg and implant the resulting embryo in a surrogate southern white rhino. "Only two embryos have ever been created," says Sieffert. "One grew to two cells and one grew to three cells but weren't viable after that. A rhino's a lot more than three cells. The technology just isn't there." In 2015, the UK nonprofit organization Protect announced it would help save rhinos by installing cameras in their horns. This system, called RAPID (Real-time Anti-Poaching Intelligence Device), would comprise the camera, a GPS collar and a heart-rate monitor. A suddenly rapid heart rate would tell the system to switch on the camera, sound an alarm and dispatch an anti-poaching team. It sounds great in theory. It also raised quite a few questions: How long would the power last? What if the rhino damaged anything? Could poachers steal or destroy the camera? Protect seems to have backed away from the idea. At the time of this writing, the organization had removed any mention of RAPID from its website and YouTube channel. And its chief architect, Paul O'Donoghue of the University of Chester, in England, didn't respond to requests for comment. Snitch took a more feasible approach. Using GPS trackers, satellite imagery and analytics software, his team created models predicting the movements of rhinos, rangers and poachers in South Africa's Olifants West Reserve. "We now have 11 months of data on every patrol route, every animal seen, every anomaly the rangers spotted," says Snitch. "I now have an organic model of how the reserve breathes, how people and animals move, so I know when and where to target poachers." But not everything has to be state of the art. Simple trickery can work, too. "I put $12 fake CCTV cameras with motion and blinking lights up in a number of trees -- along with a couple of real cameras," says Snitch, now director of federal relations at Bowling Green State University in Ohio. "The poachers think I have the fence line covered. In the area we are currently operating, we have cut poaching by 87 percent." He's also using the nature to aid efforts to supply technology in the field, an ingenious combination that relies on a firm understanding of the way the local environment works. "I am working on building beehives in Zambia equipped with solar panels, radio repeaters and a video camera on a telescopic mast," he says. "Honey to sell, better crop pollination, power to charge cell phones and lights at night, better ranger communications and the elephants stay away from village gardens -- reducing animal/human conflict. Elephants are deathly afraid of bees. All for $600 a hive. No one is going to steal my beehive." Saving the rhinoceros involves more than stopping poachers before they kill. We also have to show local communities they can gain more from the rhino's survival than they can by killing it, says IRF's Sieffert. "That can happen with ecotourism, through hiring rangers, paying local communities," she says. "We have one group in Indonesia that we pay to gather food for the animals in the sanctuary. In Tanzania, communities can create what's called communal conservancies and then sell concessions to tourism companies. That community-based conservation is a critical link in wildlife conservation." Then there's the issue of combating the illegal traffic. Both Sieffert and Snitch praise sniffer dogs, like those trained by the African Wildlife Foundation (AWF), for their role in that. Over two months, the AWF's Conservation Canine Programme trains dogs and handlers to be sent to key export hubs, such as Tanzania's port of Dar es Salaam. Sniffer dogs can detect even the smallest amounts of rhino horn dust with a 90 percent accuracy rate, the AWF claims. The first training class of eight dogs and 14 handlers -- rangers from Tanzania's Wildlife Division and the Kenya Wildlife Service -- graduated last year. Time may be running out for the rhinoceros. Greed powered by uninformed demand could wipe out rhinos in the wild in the next 20 years. But that's not stopping people from trying to save them. Last-gasp moonshots, like reconstructing one rhino species from stem cells, may yield advances to save others. GPS trackers, cameras and motion sensors could catch poachers before they can kill. Cancer breakthroughs might quash demand for the horn. There's a good chance it won't be any one answer for saving the rhino, but a combination of many. "We are excited to work with technology companies, and we hope that we can find a solution to this confounding crisis," Sieffert says. "No idea is too crazy to just throw out there and see if it actually works." About 29,000 rhinos still live in the wild. Ideas, anyone? This story appears in the winter 2016 edition of CNET Magazine. For other magazine stories, click here.
Taheri S.,Digital Signal |
Qiu Q.,Duke University |
Chellappa R.,Institute for Advanced Computer Studies
IEEE Transactions on Image Processing | Year: 2014
Although facial expressions can be decomposed in terms of action units (AUs) as suggested by the facial action coding system, there have been only a few attempts that recognize expression using AUs and their composition rules. In this paper, we propose a dictionary-based approach for facial expression analysis by decomposing expressions in terms of AUs. First, we construct an AU-dictionary using domain experts' knowledge of AUs. To incorporate the high-level knowledge regarding expression decomposition and AUs, we then perform structure-preserving sparse coding by imposing two layers of grouping over AU-dictionary atoms as well as over the test image matrix columns. We use the computed sparse code matrix for each expressive face to perform expression decomposition and recognition. Since domain experts' knowledge may not always be available for constructing an AU-dictionary, we also propose a structure-preserving dictionary learning algorithm, which we use to learn a structured dictionary as well as divide expressive faces into several semantic regions. Experimental results on publicly available expression data sets demonstrate the effectiveness of the proposed approach for facial expression analysis. © 1992-2012 IEEE.
Patro R.,Institute for Advanced Computer Studies |
Patro R.,University of Maryland University College |
Kingsford C.,Institute for Advanced Computer Studies |
Kingsford C.,University of Maryland University College
Bioinformatics | Year: 2012
Motivation: Protein interaction networks provide an important system-level view of biological processes. One of the fundamental problems in biological network analysis is the global alignment of a pair of networks, which puts the proteins of one network into correspondence with the proteins of another network in a manner that conserves their interactions while respecting other evidence of their homology. By providing a mapping between the networks of different species, alignments can be used to inform hypotheses about the functions of unannotated proteins, the existence of unobserved interactions, the evolutionary divergence between the two species and the evolution of complexes and pathways.Results: We introduce GHOST, a global pairwise network aligner that uses a novel spectral signature to measure topological similarity between subnetworks. It combines a seed-and-extend global alignment phase with a local search procedure and exceeds state-of-the-art performance on several network alignment tasks. We show that the spectral signature used by GHOST is highly discriminative, whereas the alignments it produces are also robust to experimental noise. When compared with other recent approaches, we find that GHOST is able to recover larger and more biologically significant, shared subnetworks between species. © The Author 2012. Published by Oxford University Press. All rights reserved.
Lieberman M.D.,Institute for Advanced Computer Studies |
Samet H.,Institute for Advanced Computer Studies
SIGIR'11 - Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval | Year: 2011
News sources on the Web generate constant streams of information, describing many aspects of the events that shape our world. In particular, geography plays a key role in the news, and enabling geographic retrieval of news articles involves recognizing the textual references to geographic locations (called toponyms) present in the articles, which can be difficult due to ambiguity in natural language. Toponym recognition in news is often accomplished with algorithms designed and tested around small corpora of news articles, but these static collections do not reflect the streaming nature of online news, as evidenced by poor performance in tests. In contrast, a method for toponym recognition is presented that is tuned for streaming news by leveraging a wide variety of recognition components, both rule-based and statistical. An evaluation of this method shows that it outperforms two prominent toponym recognition systems when tested on large datasets of streaming news, indicating its suitability for this domain.
Kuter U.,University of Maryland College Park |
Golbeck J.,Institute for Advanced Computer Studies
ACM Transactions on Internet Technology | Year: 2010
In this article, we describe a new approach that gives an explicit probabilistic interpretation for social networks. In particular, we focus on the observation that many existing Web-based trust-inference algorithms conflate the notions of "trust" and "confidence," and treat the amalgamation of the two concepts to compute the trust value associated with a social relationship. Unfortunately, the result of such an algorithm that merges trust and confidence is not a trust value, but rather a new variable in the inference process. Thus, it is hard to evaluate the outputs of such an algorithm in the context of trust inference. This article first describes a formal probabilistic network model for social networks that allows us to address that issue. Then we describe SUNNY, a new trust inference algorithm that uses probabilistic sampling to separately estimate trust information and our confidence in the trust estimate and use the two values in order to compute an estimate of trust based on only those information sources with the highest confidence estimates. We present an experimental evaluation of SUNNY. In our experiments, SUNNY produced more accurate trust estimates than the well-known trust inference algorithm TIDALTRUST, demonstrating its effectiveness. Finally, we discuss the implications these results will have on systems designed for personalizing content and making recommendations. © 2010 ACM.
Quercini G.,Institute for Advanced Computer Studies |
Samet H.,Institute for Advanced Computer Studies |
Sankaranarayanan J.,Institute for Advanced Computer Studies |
Lieberman M.D.,Institute for Advanced Computer Studies
GIS: Proceedings of the ACM International Symposium on Advances in Geographic Information Systems | Year: 2010
Information sources on the Internet (e.g., Web versions of newspapers) usually have an implicit spatial reader scope, which is the geographical location for which the content has been primarily produced. Knowledge of the spatial reader scope facilitates the construction of a news search engine that provides readers a set of news sources relevant to the location in which they are interested. In particular, it plays an important role in disambiguating toponyms (e.g., textual specifications of geographical locations) in news articles, as the process of selecting an interpretation for the toponym often reduces to one of selecting an interpretation that seems natural in the context of the spatial reader scope. The key to determining the spatial reader scope of news sources is the notion of local lexicon, which for a location s is a set of concepts such as, but not limited to, names of people, landmarks, and historical events, that are spatially related to s. Techniques to automatically generate the local lexicon of a location by using the link structure of Wikipedia are described and evaluated. A key contribution is the improvement of existing methods used in the semantic relatedness domain to extract concepts spatially related to a given location from the Wikipedia. Results of experiments are presented that indicate that the knowledge of the spatial reader scope significantly improves the disambiguation of textually specified locations in news articles and that using local lexicons is an effective method to determine the spatial reader scopes of news sources. © 2010 ACM.
Kelley D.R.,Institute for Advanced Computer Studies |
Kelley D.R.,University of Maryland University College |
Salzberg S.L.,Institute for Advanced Computer Studies |
Salzberg S.L.,University of Maryland University College
BMC Bioinformatics | Year: 2010
Background: Sequencing of environmental DNA (often called metagenomics) has shown tremendous potential to uncover the vast number of unknown microbes that cannot be cultured and sequenced by traditional methods. Because the output from metagenomic sequencing is a large set of reads of unknown origin, clustering reads together that were sequenced from the same species is a crucial analysis step. Many effective approaches to this task rely on sequenced genomes in public databases, but these genomes are a highly biased sample that is not necessarily representative of environments interesting to many metagenomics projects.Results: We present SCIMM (Sequence Clustering with Interpolated Markov Models), an unsupervised sequence clustering method. SCIMM achieves greater clustering accuracy than previous unsupervised approaches. We examine the limitations of unsupervised learning on complex datasets, and suggest a hybrid of SCIMM and supervised learning method Phymm called PHYSCIMM that performs better when evolutionarily close training genomes are available.Conclusions: SCIMM and PHYSCIMM are highly accurate methods to cluster metagenomic sequences. SCIMM operates entirely unsupervised, making it ideal for environments containing mostly novel microbes. PHYSCIMM uses supervised learning to improve clustering in environments containing microbial strains from well-characterized genera. SCIMM and PHYSCIMM are available open source from http://www.cbcb.umd.edu/software/scimm. © 2010 Kelley and Salzberg; licensee BioMed Central Ltd.