ACM ​Faculty Talk

Mixed Membership Word Embeddings for Computational Social Science

​Dr. James Foulds, Information Systems, UMBC

12:00-1:00pm ​Thursday​,​ 5 ​April​ 2018, ITE​459,​ UMBC

Word embeddings improve the performance of natural language processing (NLP) systems by revealing the hidden structural relationships between words. Despite their success in many applications, word embeddings have seen very little use in computational social science NLP tasks, presumably due to their reliance on big data, and to a lack of interpretability. I propose a probabilistic model-based word embedding method which can recover interpretable embeddings, without big data. The key insight is to leverage mixed membership modeling, in which global representations are shared, but individual entities (i.e., dictionary words) are free to use these representations to uniquely differing degrees. I show how to train the model using a combination of state-of-the-art training techniques for word embeddings and topic models. The experimental results show an improvement in predictive language modeling of up to 63% in MRR over the skip-gram, and demonstrate that the representations are beneficial for supervised learning. I illustrate the interpretability of the models with computational social science case studies on State of the Union addresses and NIPS articles.

James (a.k.a. Jimmy) Foulds is an assistant professor in the Department of Information Systems at UMBC. His research interests are in both applied and foundational machine learning, focusing on probabilistic latent variable models and the inference algorithms to learn them from data. His work aims to promote the practice of latent variable modeling for multidisciplinary research in areas including computational social science and the digital humanities. He earned his Ph.D. in computer science at the University of California, Irvine, and was a postdoctoral scholar at the University of California, Santa Cruz, followed by the University of California, San Diego. His master’s and bachelor’s degrees were earned with first class honours at the University of Waikato, New Zealand, where he also contributed to the Weka data mining system.

The post 🗣 talk: Mixed Membership Word Embeddings for Computational Social Science, 12pm Thr 4/5 appeared first on Department of Computer Science and Electrical Engineering.

The UMBC at Shady Grove Cybersecurity Program Presents

Cybersecurity as a Team Sport – Understanding Counsel’s Role

Allison Bender J.D.

6:00-8:00pm Wednesday, 4 April 2018

The Universities at Shady Grove
Building III (Camille Kendall Academic Center) Room 4230
9636 Gudelsky Drive, Rockville, Maryland 20850
Directions/Parking

In this presentation by seasoned incident response counsel, Allison Bender, you will gain a risk-informed perspective on the role of counsel in cybersecurity governance and incident response. Also, learn strategies for more effective communication and cooperation with other cybersecurity stakeholders (e.g., IT, IT Security, HR, Communications, business leaders, senior executives and the board); and take away practical tips for prioritizing efforts that help tame the chaos of cybersecurity incident response while maintaining privilege as appropriate.

Allison Bender counsels Fortune 50 companies and startups in a range of industries on cybersecurity and privacy matters in the U.S. and internationally. Drawing from her roots in government, national security, and R&D, she helps clients navigate legal issues associated with emerging technologies and aids clients in strategically managing legal, financial, and reputational cybersecurity risks. Allison translates technical, operational, legal, and policy issues to create practical solutions for clients’ legal challenges. Her cybersecurity and national security preparedness counseling is informed by over 80 incident response efforts. When drafting corporate policies and considering product design options, Allison’s advice is seasoned in the management of breaches involving personal data, intellectual property, payment card information, export controlled technical data, and other regulated information. Her experience also extends to counseling on cybersecurity and national security due diligence in mergers and acquisitions, vendor management, and transactions. From DHS, Allison brings experience in incident response as well as cybersecurity policy, information sharing, liability, and incentives. She was the primary operational legal counsel for the federal response to the Heartbleed vulnerability, the USIS-KeyPoint data breach, and the Healthcare.gov data breach.

Hosts: Dr. Behnam Shariati (*protected email*) and the UMBC Graduate Cybersecurity Association at USG

The UMBC Graduate Cybersecurity Association at USG is an organization created and managed by UMBC Cybersecurity graduate students at Shady Grove. The mission of Cybersecurity Association is to promote the study of Cybersecurity and to raise Cybersecurity awareness and knowledge in the community through panel discussions, conferences, and Cyber competitions. Also, the Cybersecurity Association aspires to create a supportive and positive learning environment in which every member has the opportunity to network, learn, and grow.

The post talk: Cybersecurity as a Team Sport – Understanding Counsel’s Role, 6pm Wed 4/4 appeared first on Department of Computer Science and Electrical Engineering.

UMBC ACM Student Chapter Entreprenuership Talk

Innovation and Entrepreneurship

Dr. Neil Rothman
Graduate Program Director for Professional Programs in Engineering
12.00-1.00pm Friday,​ 30 ​March 2018, ITE​​237, UMBC

There is a misconception that entrepreneurs only create start-up companies based on a new technology or product idea or that entrepreneurship is risky. We will discuss what entrepreneurship really means and how entrepreneurs can identify opportunities to create value.

Dr. Neil Rothman, Professor of the Practice and Graduate Program Director for Professional Programs in Engineering, came to UMBC as a Lecturer in 2012. Dr. Rothman has a BS in Biomedical Engineering and MS in Mechanical Engineering from Rensselaer Polytechnic Institute and a Ph.D. in Mechanical Engineering from The Johns Hopkins University. Prior to UMBC, Dr. Rothman spent over 30 years in the medical device industry, focused on product development and manufacturing, and held senior technical management positions in organizations from start-ups to large multinationals. He uses his extensive industry experience to provide a professional engineering practice perspective to all of his courses. Most recently, he was the Vice President of Advanced Research and Technology Development at BrainScope Company, Inc., leading the creation of technologies for the diagnosis of traumatic brain injury and other neurological conditions.

Prior to joining BrainScope, Dr. Rothman was the VP of Research and Development for Infinite Biomedical Technologies (IBT), and led the development of EEG based systems for detection of asphyxic brain injury and seizures in critical care applications. Other positions included Project Manager for GE Healthcare’s Maternal and Infant Care Division, VP of Operations for Metasensors (a startup company developing a microfluidic system for respiratory gas analysis), Director of Engineering for IGEN International (manufacturer of systems for high throughput drug screening, food testing, and clinical diagnostics), and Senior VP and Chief Technical Officer for CardioLogic Systems (manufacturer of automated systems for cardiopulmonary resuscitation and cardiac assist). He also held a variety of positions at the Johns Hopkins Applied Physics Laboratory and Black & Decker’s medical products division. Dr. Rothman is a Fellow of the American Society of Mechanical Engineers and recently received his 25th patent.

Contact *protected email*  with questions and follow facebook page http://goo.gl/eoMAbw for event updates

The post 🗣 talk: Innovation and Entrepreneurship, 12-1pm Fri, 3/30, ITE237 appeared first on Department of Computer Science and Electrical Engineering.

Internet of Acoustic Things (IoAT):
Challenges, Opportunities, and Threats

Nirupam Roy, University of Illinois, Urbana-Champaign

10:30-11:30am Wed. 28 March 2018, ITE325b, UMBC

The recent proliferation of acoustic devices, ranging from voice assistants to wearable health monitors, is leading to a sensing ecosystem around us — referred to as the Internet of Acoustic Things or IoAT. My research focuses on developing hardware-software building blocks that enable new capabilities for this emerging future. In this talk, I will sample some of my projects. For instance, (1) I will demonstrate carefully designed sounds that are completely inaudible to humans but recordable by all microphones. (2) I will discuss our work with physical vibrations from mobile devices, and how they conduct through finger bones to enable new modalities of short range, human-centric communication. (3) Finally, I will draw attention to various acoustic leakages and threats that arrive with sensor-rich environments. I will conclude this talk with a glimpse of my ongoing and future projects targeting a stronger convergence of sensing, computing, and communications in tomorrow’s IoT, cyber-physical systems, and healthcare technologies.

Nirupam Roy is a Ph.D. candidate in Electrical and Computer Engineering at the University of Illinois, Urbana-Champaign (UIUC). His research interests are in mobile sensing, wireless networking, and embedded systems with applications to IoT, cyber-physical-systems, and security. Roy is the recipient of the Valkenburg graduate research award, the Lalit Bahl fellowship, and the outstanding thesis awards from both his Bachelor’s and Master’s institutes. His recent research on “Making Microphones Hear Inaudible Sounds” received the MobiSys’17 best paper award and was selected for the ACM SIGMOBILE research highlights of the year in 2017.

The post 🗣talk: Internet of Acoustic Things: Challenges, Opportunities & Threats, 10:30 3/28 appeared first on Department of Computer Science and Electrical Engineering.

CHMPR Distinguished Lecture

Challenges and pitfalls in big data analysis

Yoav Benjamini, Tel Aviv University

3:30-5:00 Thursday, 12 April 2018, ITE 325b, UMBC

I shall warn about the pitfalls resulting from the false assurance that “we have all data at hand”, and discuss the challenges that are not commonly recognised such as the validity and replicability of the analysis results. Examples will be given from our work on the Health Informatics part of the European Human Brain Project, as well as from our studies in neuroscience and genomics.

Yoav Benjamini is the Nathan and Lily Silver Professor of Applied Statistics at the Department of statistics and operations research at Tel Aviv University. He holds B.Sc in physics and mathematics and M.Sc in mathematics from the Hebrew University (1976), and Ph.D in Statistics from Princeton University (1981). He is a member of the Sagol School of Neuroscience, and of the Edmond Safra Bioinformatics Center both at Tel Aviv University. He taught as a visiting professor at Wharton, UC Berkeley and Stanford and is currently visiting Columbia University. Prof. Benjamini is a co-developer of the widely used and cited False Discovery Rate concept and methodology. His current research topics are selective and simultaneous inference, replicability and reproducibility in science, model selection, and data mining. His applied research fields are Biostatistics, Bioinformatics, Animal Behavior and Brain Imaging, and as a member of the European Human Brain Project he is involved in health informatics research. Prof. Benjamini served as the president of the Israel Statistical Association, He received the Israel Prize for research in Statistics and Economics at 2012, and is an elected member of the Israel Academy of Sciences and Humanities.

The post 🗣️ talk: Challenges and pitfalls in big data analysis appeared first on Department of Computer Science and Electrical Engineering.

Computer Vision for Autonomous Underwater Vehicles

Dr. David Chapman, Oceaneering International

11:00-12:00 Monday March 12, 2018, ITE 325, UMBC

Autonomous Underwater Vehicles (AUVs) are unmanned and unteathered submarine vehicles with a variety of applications from bathymetry survey to naval warfare. Attenuation and scattering of light and electromagnetic radiation through water severely restricts wireless communications as well as distorts and attenuates camera imagery. Bandwidth limitations prevent AUVs from being remotely piloted, thus full autonomy is required for operation. Computer vision extends the ability for AUVs to perform advanced behaviors, but must address the unique challenges of underwater photography, underwater lidar, and multibeam sonar sensors. We will discuss recent research and development efforts related to computer vision of AUVs as their applications, including oilfield pipeline survey and inspection, obstacle avoidance and autonomous docking. We will also briefly discuss efforts toward amphibious vehicles, AGVs for factory automation, as well as ongoing research in acoustic signal processing.

Dr. David Chapman is a Senior Software Engineer with Oceaneering International inc., which is the largest producer of subsea Remotely Operated Vehicles (ROVs) and largest operator of Autonomous Underwater Vehicles (AUVs). Dr. Chapman completed his Ph.D. from University of Maryland Baltimore County (UMBC) in 2012 studying remote sensing, image processing, and parallel computing. He also completed a post doctoral fellowship at Columbia University’s Lamont Doherty Earth Observatory studying data analytics for El Nino prediction. At Oceaneering, Dr. Chapman has been a key contributor to computer vision algorithms research for new product development including the Pipeline Inspection AUV (PI-AUV), winner of Oceaneering’s 2017 innovative product award. He is also a contributor to both the proposal and development efforts of a vision-based AUV auto-docking system. Dr. Chapman has studied and applied a variety of computer vision algorithms including the fast Radon transform, wavelet-based feature classification, numerical optimization, and neural networks in order to extend the capabilities of AUVs and related autonomous vehicles.

The post 🗣️talk: Computer Vision for Autonomous Underwater Vehicles, 11am Mon 3/12 appeared first on Department of Computer Science and Electrical Engineering.

The UMBC Cyber Defense Lab presents

   Creating Educational Cybersecurity Assessment Tools

Alan T. Sherman
Department of Computer Science and Electrical Engineering
University of Maryland, Baltimore County

12:00–1:00pm Friday, March 9, 2018, ITE 229, UMBC

The Cybersecurity Assessment Tools (CATS) Project provides rigorous evidence-based instruments for assessing and evaluating educational practices. The first CAT will be a Cybersecurity Concept Inventory (CCI) that measures how well students understand basic concepts in cybersecurity (especially adversarial thinking) after a first course in the field. The second CAT will be a Cybersecurity Curriculum Assessment (CCA) that measures how well students understand core concepts after completing a full cybersecurity curriculum. These tools can help identify pedagogies and content that are effective in teaching cybersecurity.

In fall 2014, we carried out a Delphi process that identified core concepts of cybersecurity. In spring 2016, we interviewed twenty-six students to uncover their understandings and misconceptions about these concepts. In fall 2016, we generated our first assessment tool—-a draft CCI, comprising approximately thirty multiple-choice questions. Each question targets a concept; incorrect answers are based on observed misconceptions from the interviews. In fall 2017, we began drafting CCA questions. This year we are validating the draft CCI using cognitive interviews, expert reviews, and psychometric testing. In this talk, I highlight our progress to date in developing the CCI and CCA. Audience members will be given an opportunity to answer sample questions.

Presently there is no rigorous, research-based method for measuring the quality of cybersecurity instruction. Validated assessment tools are needed so that cybersecurity educators have trusted methods for discerning whether efforts to improve student preparation are successful.

Joint work with Linda Oliva, David DeLatte, Enis Golaszewski, Geet Parekh, Konstantinos Patsourakos, Dhananjay Phatak, Travis Scheponik (UMBC); Geoffrey Herman, Dong San Choi, Julia Thompson (University of Illinois at Urbana-Champaign)

Alan T. Sherman is a professor of computer science at UMBC in the CSEE Department and Director of UMBC’s Center for Information Security and Assurance. His main research interest is high-integrity voting systems. He has carried out research in election systems, algorithm design, cryptanalysis, theoretical foundations for cryptography, applications of cryptography, and cybersecurity education. Dr. Sherman is also an editor for Cryptologia and a private consultant performing security analyses. Sherman earned the PhD degree in computer science at MIT in 1987 studying under Ronald L. Rivest. www.csee.umbc.edu/~sherman

Support for this research was provided in part by the National Security Agency under grants H98230-15-1-0294 and H98230-15-1-0273 and by the National Science Foundation under SFS grant 1241576.

The post talk: Creating Educational Cybersecurity Assessment Tools, 12pm Fri 3/9 appeared first on Department of Computer Science and Electrical Engineering.

UMBC ACM Student Chapter

Planning and Learning in Complex Stochastic Domains: AMDPs, Option Discovery, Learning Transfer, Language Learning, and More

Dr. Marie desJardins, University of Maryland, Baltimore County
1-2pm Friday, March 9th, 2018, ITE 456, UMBC

Robots acting in human-scale environments must plan under uncertainty in large state–action spaces and face constantly changing reward functions as requirements and goals change. We introduce a new hierarchical planning framework called Abstract Markov Decision Processes (AMDPs) that can plan in a fraction of the time needed for complex decision making in ordinary MDPs. AMDPs provide abstract states, actions, and transition dynamics in multiple layers above a base-level “flat” MDP. AMDPs decompose problems into a series of subtasks with both local reward and local transition functions used to create policies for subtasks. The resulting hierarchical planning method is independently optimal at each level of abstraction, and is recursively optimal when the local reward and transition functions are correct.

I will present empirical results in several domains showing significantly improved planning speed, while maintaining solution quality. I will also discuss related work within the same project on automated option discovery, abstraction construction, language learning, and initial steps towards automated methods for learning AMDPs from base MDPs, from teacher demonstrations, and from direct observations in the domain.

This work is collaborative research with Dr. Michael Littman and Dr. Stefanie Tellex of Brown University. Dr. James MacGlashan of SIFT and Dr. Smaranda Muresan of Columbia University collaborated on earlier stages of the project. The following UMBC students have also contributed to the project: Khalil Anderson, Tadewos Bellete, Michael Bishoff, Rose Carignan, Nick Haltemeyer, Nathaniel Lam, Matthew Landen, Keith McNamara, Stephanie Milani, Shane Parr (UMass), Shawn Squire, Tenji Tembo, Nicholay Topin, Puja Trivedi, and John Winder.

Dr. Marie desJardins is a Professor of Computer Science and the Associate Dean for Academic Affairs in the College of Engineering and Information Technology at the University of Maryland, Baltimore County. Prior to joining the faculty at UMBC in 2001, she was a Senior Computer Scientist in the AI Center at SRI International. Her research is in artificial intelligence, focusing on the areas of machine learning, multi-agent systems, planning, interactive AI techniques, information management, reasoning with uncertainty, and decision theory. She is active in the computer science education community, founded the Maryland Center for Computing Education, and leads the CS Matters in Maryland project to develop curriculum and train high school teachers to teach AP CS Principles.

Dr. desJardins has published over 125 scientific papers in journals, conferences, and workshops. She will be the IJCAI-20 Conference Chair, and has been an Associate Editor of the Journal of Artificial Intelligence Research and the Journal of Autonomous Agents and Multi-Agent Systems, a member of the editorial board of AI Magazine, and Program Co-chair for AAAI-13. She has previously served as AAAI Liaison to the Board of Directors of the Computing Research Association, Vice-Chair of ACM’s SIGART, and AAAI Councillor. She is a AAAI Fellow, an ACM Distinguished Member, a Member-at-Large for Section T (Information, Computing, and Communication) of the American Association for the Advancement of Science, the 2014-17 UMBC Presidential Teaching Professor, a member and former chair of UMBC’s Honors College Advisory Board, former chair of UMBC’s Faculty Affairs Committee, and a member of the advisory board of UMBC’s Center for Women in Technology.

The post talk: desJardins on Planning and Learning in Complex Stochastic Domains, 1pm fri 3/8 appeared first on Department of Computer Science and Electrical Engineering.

The UMBC Cyber Defense Lab presents

Experimentally Measuring the Circuit Complexity
of One-Way Boolean Functions

Brian Weber, CSEE, UMBC

12:00–1:00pm, Friday, 23 February 2018, ITE 229

I present preliminary results from an exhaustive search for one-way functions in certain classes of small Boolean functions.   One-way functions are functions that are easy to compute but hard to invert.  They are vital for cryptography, yet no one has proven their existence for arbitrary input sizes.  For any bounded circuit model of computation, it is possible to search exhaustively over all possible Boolean functions of restricted size and thereby determine for the searched class the maximum disparity between the complexity of any function and its inverse.  Throughout, we assume a circuit model in which each gate has fan-in 2 and fan-out 1.

In his 1985 dissertation at MIT, Steven Boyack carried out the first such search.  For any positive integers n and M, let F_n,M denote the set of Boolean functions with n inputs and Moutputs. Using circuit size as the complexity measure, Boyack searched the space of every combinatorial function in F_3,3 by searching each of 52 equivalency classes of functions in this space.  He found that every function class in this space has an identically sized inverse.  He was able to prove that functions do exist with more complex inverses outside the space he searched, but not by more than a constant factor.

In spring 2017, using circuit depth as the complexity measure, I searched all injective functions up to F_8,8 whose coordinate functions are in F_2,1.  A coordinate function in this context refers to the function that computes an individual output bit.  In addition, I searched up to F_4,4 allowing coordinate functions in F_3,1.  In the space I searched, the most one-way function has fixed depth of 1, and an inverse depth exactly equal to the input size of the function. That is, for each 2 < n < 9, the hardest inverse in the space I searched has a depth of n, where n is the number of input bits. In addition, a search space allowing a larger fan-in for the coordinate functions did not yield functions less invertible than were found in the original search space.

Brian Weber is a senior BS/MS computer engineering student and SFS scholar at UMBC.  He hopes to extend the work presented here into his Master’s thesis next year.  Email: *protected email*

Host: Alan T.  Sherman, *protected email*Support for this research was provided in part by the National Science Foundation under SFS grant 1241576.

The UMBC Cyber Defense Lab meets biweekly Fridays.  All meetings are open to the public.

The post talk: Circuit Complexity of One-Way Boolean Functions, 12pm Fri 2/23, ITE229 appeared first on Department of Computer Science and Electrical Engineering.

ACM Faculty Talk Series

Semi-supervised Learning for Visual Recognition

Dr. Hamed Pirsiavash, Assistant Professor, CSEE

1:00-2:00pm Friday, February 23, 2018, ITE 325, UMBC

We are interested in learning representations (features) that are discriminative for semantic image understanding tasks such as object classification, detection, and segmentation in images. A common approach to obtain such features is to use supervised learning. However, this requires manual annotation of images, which is costly, time-consuming, and prone to errors. In contrast, unsupervised or self-supervised feature learning methods exploiting unlabeled data can be much more scalable and flexible. I will present some of our efforts in this direction.

Hamed Pirsiavash is an assistant professor at the University of Maryland, Baltimore County (UMBC). Prior to joining UMBC in 2015 he was a postdoctoral research associate at MIT and he obtained his PhD at the University of California Irvine. He does research in the intersection of computer vision and machine learning.

This talk is sponsored by the UMBC Student Chapter of the ACM. Contact *protected email* with any questions regarding this event.

The post talk: Semi-supervised Learning for Visual Recognition, 1pm Fri 2/23, ITE325, UMBC appeared first on Department of Computer Science and Electrical Engineering.