News Article | July 24, 2017
NEW YORK--(BUSINESS WIRE)--Citigroup Inc. (“Citi”) today announced a collaboration with Cornell Tech, the revolutionary graduate institution, to engage with Cornell University students, faculty, researchers, startups and other companies, and create closer ties between academia and industry. The collaboration will be anchored by Citi’s presence on the Cornell Tech campus on Roosevelt Island. Citi has leased 10,900 square feet in The Bridge at Cornell Tech from Forest City New York and is the only bank located in the building located on Roosevelt Island. The Bridge is a first-of-its-kind building for innovation that brings together residents of the Cornell Tech campus, to foster a culture of collaboration and co-creation. “Co-locating in The Bridge with companies, entrepreneurs and students will create an energized environment that fosters innovation and enables interaction with a wide range of emerging talent,” said Don Callahan, Citi’s Head of Operations and Technology. “As a firm headquartered in New York City, we are excited to be a part of this initiative with Cornell Tech that will develop and strengthen the City’s technology talent and industry.” Citi will tap into the expertise of Cornell Tech students and faculty to work closely on new capabilities and emerging technologies such as blockchain, machine learning and big data applications, biometric authentication, Internet of Things, and cyber security. The partnership will provide faculty and students the opportunity to work with Citi in exploring real-world solutions for Citi’s clients and customers, through Product Challenges in Cornell Tech’s Product Studio class, where student teams develop innovative new digital solutions working with companies and nonprofits. Through programs such as the Citi Ventures University Partnerships program, Citi teams will engage with Cornell Tech students in activities such as hackathons and design sprints to create, test and incubate new ideas and prototypes that have the potential to generate organic growth for Citi. In addition, the Citi Foundation will continue to work with Cornell Tech through their mutual support of the Women in Technology and Entrepreneurship in New York (WiTNYC) initiative, a program designed to significantly increase the participation of women in technology fields in the New York market. Citi will accommodate approximately 70 to 80 employees in this facility. These employees are responsible for advancing the bank’s customer banking experience, partnering with start-ups on innovative technologies and safeguarding clients’ information. The new flexible work environment will drive collaboration between Citi employees, Cornell Tech students and faculty, and other corporate tenants. Citi employees will move into The Bridge during the first quarter of 2018. “Bringing together industry leaders like Citi and our pioneering students and faculty focusing on key problems and opportunities of the digital age is part of Cornell Tech’s lifeblood,” said Cornell Tech Dean Dan Huttenlocher. “Collaborations like this one will bring The Bridge and the entire campus to life when we open on Roosevelt Island this fall.” “Dedicated to uncovering emerging technologies and creative results for their clients and customers, Citi is a great fit for the dynamic environment at The Bridge at Cornell Tech,” said MaryAnne Gilmartin, President and CEO of Forest City New York. “All under one roof, Citi will have easy access to Cornell University students and faculty, inspiring deep innovation.” Citi’s presence in New York City includes multiple office locations, its branch network and nearly 16,000 employees. CBRE’s Mary Ann Tighe, Evan Haskell, David Caperna, Evan Fiddle, Sacha Zarba and Ross Zimbalist brokered the lease on behalf of Forest City New York. About Citi Citi, the leading global bank, has approximately 200 million customer accounts and does business in more than 160 countries and jurisdictions. Citi provides consumers, corporations, governments and institutions with a broad range of financial products and services, including consumer banking and credit, corporate and investment banking, securities brokerage, transaction services, and wealth management. About Cornell Tech Cornell Tech brings together faculty, business leaders, tech entrepreneurs, and students in a catalytic environment to reinvent the way we live in the digital age. Cornell Tech’s temporary campus has been up and running at Google’s Chelsea building since 2013, with a growing world-class faculty, and more than 200 masters and Ph.D. students who collaborate extensively with tech-oriented companies and organizations and pursue their own start-ups. Construction is underway on Cornell Tech’s campus on Roosevelt Island, with a first phase due to open in September 2017. When fully completed, the campus will include 2 million square feet of state-of-the-art buildings, over 2 acres of open space, and will be home to more than 2,000 graduate students and hundreds of faculty and staff. About Forest City New York Forest City New York, a wholly owned subsidiary of Forest City Realty Trust, Inc., is the owner and developer of The Bridge at Cornell Tech, and owns and operates over 30 properties in the New York metropolitan area, including The New York Times Building. Forest City Realty Trust, Inc. is an NYSE-listed national real estate company with $8.2 billion in consolidated assets. The Company is principally engaged in the ownership, development, management and acquisition of commercial and residential real estate throughout the United States, and is the developer of such projects as University Park at MIT, the Science + Technology Park at Johns Hopkins, and 5M in San Francisco. For more information, visit www.forestcity.net. About Cornell University Cornell University is a world-class research institution known for the breadth and rigor of its curricula, and an academic culture dedicated to preparing students to be well-educated and well-rounded citizens of the world. Its faculty, staff and students believe in the critical importance of knowledge—both theoretical and applied—as a means of improving the human condition and solving the world’s problems. With campuses in Ithaca, New York, New York City, and Doha, Qatar, Cornell is a private, Ivy League research university and the land-grant institution of New York state.
Sun E.,Cornell Technology
CSCW 2017 - Companion of the 2017 ACM Conference on Computer Supported Cooperative Work and Social Computing | Year: 2017
Through my doctoral research, I aim to gain a deeper understanding of how LBPHDs, location-based, post-hoc data applications, can be used to build social capital in urban communities. From a case study of a dating app that utilizes shared location history, happn, we explored how LBPHD information has been used to build interpersonal relationships. Based on these findings, we designed Move-Meant, an application that extends LBPHD from interpersonal to community-level information sharing. Preliminary qualitative field results suggest the potential of MoveMeant to increase local community awareness through dissemination of local knowledge and discovery of third places. Future work in this area include a larger qualitative study and quantitative study of MoveMeant, and further understanding its application to other situations.
Bindschaedler V.,University of Illinois at Urbana - Champaign |
Shokri R.,Cornell Technology |
Gunter C.A.,University of Illinois at Urbana - Champaign
Proceedings of the VLDB Endowment | Year: 2016
Releasing full data records is one of the most challenging problems in data privacy. On the one hand, many of the popular techniques such as data de-identification are problematic because of their dependence on the background knowledge of adversaries. On the other hand, rigorous methods such as the exponential mechanism for differential privacy are often computationally impractical to use for releasing high dimensional data or cannot preserve high utility of original data due to their extensive data perturbation. This paper presents a criterion called plausible deniability that provides a formal privacy guarantee, notably for releasing sensitive datasets: An output record can be released only if a certain amount of input records are indistinguishable, up to a privacy parameter. This notion does not depend on the background knowledge of an adversary. Also, it can efficiently be checked by privacy tests. We present mechanisms to generate synthetic datasets with similar statistical properties to the input data and the same format. We study this technique both theoretically and experimentally. A key theoretical result shows that, with proper randomization, the plausible deniability mechanism generates differentially private synthetic data. We demonstrate the efficiency of this generative technique on a large dataset; it is shown to preserve the utility of original data with respect to various statistical analysis and machine learning measures. © 2017. VLDB Endowment.
Bindschaedler V.,University of Illinois at Urbana - Champaign |
Shokri R.,Cornell Technology
Proceedings - 2016 IEEE Symposium on Security and Privacy, SP 2016 | Year: 2016
Camouflaging user's actual location with fakes is a prevalent obfuscation technique for protecting location privacy. We show that the protection mechanisms based on the existing (ad hoc) techniques for generating fake locations are easily broken by inference attacks. They are also detrimental to many utility functions, as they fail to credibly imitate the mobility of living people. This paper introduces a systematic approach to synthesizing plausible location traces. We propose metrics that capture both geographic and semantic features of real location traces. Based on these statistical metrics, we design a privacy-preserving generative model to synthesize location traces which are plausible to be trajectories of some individuals with consistent lifestyles and meaningful mobilities. Using a state-of-the-art quantitative framework, we show that our synthetic traces can significantly paralyze location inference attacks. We also show that these fake traces have many useful statistical features in common with real traces, thus can be used in many geo-data analysis tasks. We guarantee that the process of generating synthetic traces itself is privacy preserving and ensures plausible deniability. Thus, although the crafted traces statistically resemble human mobility, they do not leak significant information about any particular individual whose data is used in the synthesis process. © 2016 IEEE.
Merolla P.A.,IBM |
Arthur J.V.,IBM |
Alvarez-Icaza R.,IBM |
Cassidy A.S.,IBM |
And 16 more authors.
Science | Year: 2014
Inspired by the brain's structure, we have developed an efficient, scalable, and flexible non-von Neumann architecture that leverages contemporary silicon technology. To demonstrate, we built a 5.4-billion-transistor chip with 4096 neurosynaptic cores interconnected via an intrachip network that integrates 1 million programmable spiking neurons and 256 million configurable synapses. Chips can be tiled in two dimensions via an interchip communication interface, seamlessly scaling the architecture to a cortexlike sheet of arbitrary size. The architecture is well suited to many applications that use complex neural networks in real time, for example, multiobject detection and classification. With 400-pixel-by-240-pixel video input at 30 frames per second, the chip consumes 63 milliwatts.
Lingel J.,Microsoft |
Naaman M.,Cornell Technology |
Proceedings of the ACM Conference on Computer Supported Cooperative Work, CSCW | Year: 2014
We use qualitative interviews with 26 transnational migrants in New York City to analyze socio-technical practices related to online identity work. We focus specifically on the use of Facebook, where benefits included keeping in touch with friends and family abroad and documenting everyday urban life. At the same time, many participants also reported experiences of fatigue, socio-cultural tensions and concerns about maintaining a sense of personal privacy. These experiences highlight how transnational practices complicate context collapse, where the geographic dispersal of participants' personal networks renders visible conflicts of "flattened" online networks. Our findings also suggest a kind of technology-enabled codeswitching, where transnational migrants leverage social media to perform identities that alternate between communities, nationalities and geographies. This analysis informs HCI research on transnationalism and technological practices, and helps expose the complexities of online identity work in terms of shifting social and spatial contexts. Copyright © 2014 ACM.
Shokri R.,University of Texas at Austin |
Shmatikov V.,Cornell Technology
Proceedings of the ACM Conference on Computer and Communications Security | Year: 2015
Deep learning based on artificial neural networks is a very popular approach to modeling, classifying, and recognizing complex data such as images, speech, and text. The unprecedented accuracy of deep learning methods has turned them into the foundation of new AI-based services on the Internet. Commercial companies that collect user data on a large scale have been the main beneficiaries of this trend since the success of deep learning techniques is directly proportional to the amount of data available for training. Massive data collection required for deep learning presents obvious privacy issues. Users' personal, highly sensitive data such as photos and voice recordings is kept indefinitely by the companies that collect it. Users can neither delete it, nor restrict the purposes for which it is used. Furthermore, centrally kept data is subject to legal subpoenas and extra-judicial surveillance. Many data owners-for example, medical institutions that may want to apply deep learning methods to clinical records-are prevented by privacy and confidentiality concerns from sharing the data and thus benefitting from large-scale deep learning. In this paper, we design, implement, and evaluate a practical system that enables multiple parties to jointly learn an accurate neuralnetwork model for a given objective without sharing their input datasets. We exploit the fact that the optimization algorithms used in modern deep learning, namely, those based on stochastic gradient descent, can be parallelized and executed asynchronously. Our system lets participants train independently on their own datasets and selectively share small subsets of their models' key parameters during training. This offers an attractive point in the utility/privacy tradeoff space: participants preserve the privacy of their respective data while still benefitting from other participants' models and thus boosting their learning accuracy beyond what is achievable solely on their own inputs. We demonstrate the accuracy of our privacypreserving deep learning on benchmark datasets.
Juels A.,Cornell Technology
Proceedings of ACM Symposium on Access Control Models and Technologies, SACMAT | Year: 2014
Decoy objects, often labeled in computer security with the term honey, are a powerful tool for compromise detection and mitigation. There has been little exploration of overarching theories or set of principles or properties, however. This short paper (and accompanying keynote talk) briefly explore two properties of honey systems, indistinguishability and secrecy. The aim is to illuminate a broad design space that might encompass a wide array of areas in information security, including access control, the main topic of this symposium.
Pass R.,Cornell Technology
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2016
Cryptographic notions of knowledge consider the knowledge obtained, or possessed, by computationally-bounded agents under adversarial conditions. In this talk, we will survey some recent cryptographically-inspired approaches for reasoning about agents in the context of game-theory and mechanism design (where agents typically are modelled as computationally unbounded). © Springer International Publishing Switzerland 2016.
Shmatikov V.,Cornell Technology
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015
SSL/TLS is the de facto standard for secure Internet communications. Deployed widely in Web browsers and non-browser software, it is intended to provide end-to-end security even against active, man-in-the-middle attacks. This security fundamentally depends on correct validation of X.509 certificates presented when the connection is established. I will first demonstrate that many SSL/TLS deployments are completely insecure against man-in-the-middle attacks. Vulnerable software includes cloud computing clients, merchant SDKs responsible for transmitting payment information from e-commerce sites to payment processors, online shopping software, and many forms of middleware. Even worse, several popular SSL/TLS implementations do not validate certificates correctly and thus all software based on them is generically insecure. These bugs affect even common Web browsers, where minor validation errors such as recent certificate expiration can mask serious issues such as failure to authenticate the Web server’s identity. I will then analyze the root causes of these vulnerabilities and describe how we used “frankencerts,” a new methodology for automatically testing SSL/TLS implementations, to uncover dozens of subtle certificate validation bugs in popular SSL/TLS implementations. © Springer International Publishing Switzerland 2015.