banner
CS Colloquium (BMAC)
 

The Department of Computer Science of Colorado State University, in cooperation with ISTeC (Information Science and Technology Center), offers the CS Colloquium series as a service to all who are interested in computer science. When in-person meetings are possible, most seminars are scheduled for Monday 11:00AM -- 11:50AM in CSB 130 or Morgan Library Event Hall. For help finding the locations of our seminar meetings, consult the on-line CSU campus map.map

For questions about this page or to schedule talks, please contact Louis-Noel Pouchet (pouchet AT colostate dot edu). Here is a list of past seminar schedules.

CS692 information for students is available directly on Canvas.

 

Upcoming Events





CS Colloquium Schedule, Spring 2022



January
24

cs Computer Science Department Colloquium
CS Faculty Rapid-Fire Presentations of Current Research: Q&A, first session

Speaker: Computer Science Faculty, Colorado State University

When: 11:00AM ~ 11:50AM, Monday January 24, 2022
Where: CSB 130 map

Abstract: CS Faculty present and answer questions about their research, and opportunities in their research group for students at CSU.




January
31

cs Computer Science Department Colloquium
CS Faculty Rapid-Fire Presentations of Current Research: Q&A, second session

Speaker: Computer Science Faculty, Colorado State University

When: 11:00AM ~ 11:50AM, Monday January 31, 2022
Where: CSB 130 map

Abstract: CS Faculty present and answer questions about their research, and opportunities in their research group for students at CSU.




February
7

cs Computer Science Department Colloquium
Program Equivalence is More Decidable than You Think

Speaker: Louis-Noel Pouchet, Associate Professor, Department of Computer Science, Colorado State University

When: 11:00AM ~ 11:50AM, Monday February 7, 2022
Where: CSB 130 map

Abstract: Program equivalence is the process of determining when, for any possible inputs, two programs will necessarily compute the exact same outputs. It is a fundamental problem of computer science, with applications including hardware design verification and compiler validation. Taking two arbitrary programs, in general, the problem of determining their equivalence is not decidable: we cannot conclude the two programs are equivalent or not. But, there is a wide class of programs for which we can still provably determine equivalence.

In this talk we will discuss some existing approaches to prove the equivalence of two restricted programs, illustrating how increasingly complex concepts from the input language (e.g., type of loops, local variables, etc.) can be supported for automatic program equivalence computation. In particular, we will build together a simple but very powerful system for program equivalence by composing well-known concepts from various areas of computer science, inspired from our latest work on generalized equivalence by abstract evaluation. This work is funded by Intel to help them validate that optimizations of a hardware design (e.g., of a processor) have not modified the semantics of the original unoptimized design. We will also briefly discuss some deep learning techniques to automatically prove equivalence of two complex expressions.

Bio: Louis-Noel Pouchet is an Associate Professor of Computer Science at Colorado State University, with a joint appointment in the Electrical and Computer Engineering department at CSU. Prior to joining CSU in 2016, he received his Ph.D. in Computer Science in 2010 from University of Paris-Sud 11 and INRIA, France, was a postdoctoral fellow (2010-2012) and then Research Assistant Professor (2014-2016) in the Computer Science and Engineering department at Ohio State University, and was a Visiting Assistant Professor at University of California Los Angeles (2012-2014). He works on a variety of topics for high-performance computing, including optimizing compilers for CPUs, GPUs and FPGAs; domain-specific languages and optimizations; high-level synthesis for accelerator deisgn; and distributed systems for large-scale scientific computing. Pouchet is a leading expert in polyhedral compilation, a framework for loop transformations that has made its way in production compilers such as LLVM, GCC, IBM ⁄ XL, etc. His research is funded by the U.S. National Science Foundation, the U.S. Department of Energy and Intel. He has co-authored 70+ peer-reviewed publications; he is the recipient of an NSF CAREER grant, the William J. McCalla IEEE best paper award at ICCAD’15, and has received several best paper awards at top conferences. He is the main author of the PoCC polyhedral compiler, of the ROSE ⁄ PolyOpt software toolset (a complete polyhedral compilation framework for the LLNL ROSE compiler), and of the popular PolyBench ⁄ C benchmarking suite, which has been used in 400+ publications so far.




February
14

cs Computer Science Department Colloquium
Explainable Human-AI Interaction for Sequential Decision Support

Speaker: Dr. Sarath Sreedharan

When: 11:00AM ~ 11:50AM, Monday February 14, 2022
Where: CSB 130 map

Abstract: For AI-powered systems to be truly successful in practical and everyday scenarios, it is not just enough for these systems to generate optimized decisions, but they need to be capable of working and collaborating with users from all walks of life. One major requirement for developing such systems is the need to imbue them with the ability to effectively model the expectations of their users and to be capable of explaining their decisions and the rationale behind them in intuitive terms when such expectations cannot be met. In this talk, I will be describing some of the work I have done in these directions, particularly to generate explanations in various sequential decision-making settings. I will discuss my solutions to some of the core challenges related to explanation generation, namely knowledge, inferential capability, and vocabulary asymmetry between the user and the decision-maker. I will also describe some of the use-case scenarios that have leveraged these methods and end the talk by discussing how we can extend these principles to enable people to provide advice and preferences to the AI system.




February
25

cs Computer Science Department Colloquium
Bridging Programmers' Collaboration Gaps with Intelligent Programmer-Centric Systems

Speaker: Dr. Sandeep Kuttal, University of Tulsa, OK

When: 10:00AM ~ 10:50AM, Friday February 25, 2022
Where: CSB 130 map

Abstract: Software applications are developed in collaboration with different team members, yet few innovations in development tools consider human factors such as communication styles, creativity strategies, foraging behavior, and gender differences. In this talk, I will discuss my efforts to bridge this gap. I will focus on my ongoing work to create an inclusive pair programming conversational agent to foster collaborations with a human partner, including (1) my approach and design of an anthropomorphic programming agent, (2) empirical results from mixed-method studies that demonstrate the benefits of pair programming when working with an agent vs. another programmer, (3) evaluation results that demonstrate the effectiveness of machine learning algorithms for understanding programmer conversations. I will also talk about my vision, current, and future projects focusing on collective foraging intelligence of expert programmers, same- and mixed-gendered pair programmers, and tools for end-user programmers.

Bio: Sandeep Kaur Kuttal is an Assistant Professor at The University of Tulsa, where she directs the Human-Centric Software Engineering Lab. Her research combines Human-Computer Interaction, Software Engineering, and Artificial Intelligence. She focuses on the human aspects of software engineering by studying and modeling programmer behavior and then designing and developing mixed-initiative programmer-computer systems. Sandeep is a recent recipient of the NSF CAREER and AFOSR YIP awards. She has received a best paper award at ACM CHI, best paper at ACM ⁄ IEEE ICGSE, and an honorable mention at ACM CHI. She is on the editorial board of the Journal of Computer Languages, vice-chair of the steering committee for the IEEE Symposium on Visual Languages and Human-Centric Computing (VL ⁄ HCC), and acted as chair or program committee member at CHI, VL ⁄ HCC, HCII, ICSE, FSE, ASE, ICST, ICSME, CHASE, ICPC, and IUI. She is passionate about diversity and inclusion. (http: ⁄ ⁄ sandeepkuttal.ens.utulsa.edu ⁄ ).




March
3

cs Computer Science Department Colloquium
The complexity of approximate counting problems

Speaker: Dr. Ewan Davies, Computer Science Department, CU Boulder

When: 10:00AM ~ 10:50AM, Thursday March 3, 2022
Where: CSB 130 map

Abstract: Counting is a fundamental mathematical task that appears in many forms and has many applications. In theoretical computer science, we study the complexity of counting feasible solutions to constraint satisfaction problems and hope to design practical algorithms that can be used in fields such as machine learning and physics. Counting versions of classic NP-hard optimization problems such as max-cut and max independent set connect complexity theory, combinatorial optimization, graph theory and statistical physics. For example, we now have algorithms inspired by physics, physical results inspired by algorithms, and powerful mathematical tools thanks to this interdisciplinary work. My research combines the questions, intuition and techniques that permeate these topics.

In this talk we will survey approximate counting, describe some of the major results in the area, and discuss long-standing open problems. In particular, we will look at some new results connecting counting to optimization, and a particularly elegant complexity phenomenon which exemplifies the benefits of interdisciplinary thinking. This will take us through some basic physics, probability, and graph theory in order to answer a simple counting question: when can we efficiently approximate the number of independent sets of a given size in some given graph?

Bio: Ewan Davies is a postdoctoral researcher in the Department of Computer Science at the University of Colorado Boulder. His primary interests are in the complexity of approximate counting problems, which means using a wide range of mathematical techniques in order to design and analyze algorithms, and study the limits of what computers can be persuaded to do efficiently. Prior to working at CU Boulder, he studied at the University of Cambridge and received a PhD in graph theory from the London School of Economics. He has worked as a postdoc with Lex Schrijver at the University of Amsterdam and was a research fellow at the Simons Institute for the Theory of Computing for a program on the geometry of polynomials. Ewan has a background in extremal graph theory and has applied this knowledge to theoretical computer science that has been published in leading conferences such as STOC, ICALP and CCC.




March
7

cs Computer Science Department Colloquium
Improving Software Quality Using Natural Language Artifacts

Speaker: Manish Motwani, Manning College of Information & Computer Sciences, University of Massachusetts Amherst

When: 11:00AM ~ 11:50AM, Monday March 7, 2022
Where: CSB 130 map

Abstract: Software is ubiquitous but often faulty. Software engineers spend 35–50% of their time debugging software, accounting for 50–75% of the total software development budget. In this talk, I will describe my research on using natural language software artifacts to automatically improve software quality. I will first present Swami, my technique for automatically generating tests with oracles from software specifications, and, second, my method for using developer-written tests together with bug reports to automatically localize and repair bugs. I will then describe my research vision for automating software quality improvement.

Bio: Manish Motwani is a PhD candidate in Computer Science at the University of Massachusetts Amherst, advised by Prof. Yuriy Brun. Before starting his PhD, he worked for four years at the Tata Research Development and Design Center, an industrial research lab in India. His interests lie in improving software engineers’ productivity and aiding software engineers in developing high-quality software by automating software engineering practices. His research involves learning phenomena from large software repositories and using that knowledge to design novel automation techniques, such as requirements elicitation, test synthesis, and program repair.




March
10

cs Computer Science Department Colloquium
Overcoming the Fear of Needles: Why We Must Be Concerned About CAN Injection

Speaker: Mert Pesé, University of Michigan

When: 10:00AM ~ 10:50AM, Thursday March 10, 2022
Where: CSB 130 map

Abstract: The Controller Area Network (CAN) was developed in the 1980s to drive forward digitalization and data exchange in passenger vehicles and is still the predominant bus technology today. However, no one anticipated the need for security when it was developed. There are no provisions for guaranteeing sender authenticity and integrity, and all CAN messages are broadcast in plaintext. As a result, the last decade witnessed many increasingly sophisticated cyber-attacks on cars. In order to make the car malfunction ⁄ fail, all attacks have one crucial step in common: The injection of well-formed malicious CAN messages into the in-vehicle network which is a two-step process. First, the attacker has to learn the semantics of the CAN payload to inject to have a visible outcome, e.g., the car driving into a ditch. The semantics differ from car to car. Second, with the current CAN implementation, any rogue node can then easily tap into the CAN bus and inject arbitrary messages.

So far, virtually all research on protecting the CAN bus has focused on solving the problem of sender authentication using cryptographic means known from traditional network security. The first step of obtaining the semantics is mostly overlooked and can only be learned through a tedious process of manual reverse engineering. To accelerate the first step, I developed an automated CAN reverse engineering tool called LibreCAN which provides information about the target vehicle’s semantics in less than 2 minutes. By using LibreCAN for reconnaissance, I showed an apparent vulnerability in production vehicles and why confidentiality is a security property that has to be protected. To prevent eavesdropping attacks and thus reverse engineering of CAN payloads, I then developed S2-CAN. The latter also provides authenticity and integrity protection, all while meeting the resource and cost constraints that hinder the adoption of existing CAN security solutions proposed in the past decade.

Bio: Mert D. Pesé is a Ph.D. Candidate at the University of Michigan. He is advised by Professor Kang Shin and is a member of the Real Time Computing Lab (RTCL). He earned two B.S. degrees in Electrical Engineering and Computer Science respectively, and an M.S. in Electrical Engineering from the Technical University of Munich (TUM). His research interests primarily include the security and privacy of vehicles, including CAN bus, V2X, Android Automotive and adversarial machine learning for autonomous vehicles. He worked on Automotive Ethernet Security and Intrusion Detection Systems during industry positions he has held before his PhD. He has several peer-reviewed publications in top security and privacy venues, such as CCS, PETS and ACSAC.




March
21

cs Computer Science Department Colloquium
Distributed ledgers: How they work, how do we know that they work?

Speaker: Richard Brooks, Professor of Electrical and Computer Engineering, Clemson University, South Carolina.

When: 11:00AM ~ 11:50AM, Monday March 21, 2022
Where: CSB 130 map

Abstract: Distributed ledgers, the technology behind crypto-currencies, are controversial in the computer and network security community. Many luminaries are very disparaging about the utility of this innovation. This presentation looks at the technical under-pinning of Distributed Ledger Technology (DLT). DLT’s main innovation is creating a “top-down” basis for system security relying on a distributed data structure, rather than “bottom-up” security relying on a strongly defended server fortress. The presentation explains the security goals attained by explaining many of the current systems. It then looks at how to analyze the ability of these distributed systems to assure security. We concentrate specifically on consensus protocols. Explanations are given of Proof-of-Work and Proof-of-Stake. We then explain a simple, efficient light-weight mining (LWM) approach. LWM is well suited to abstract analysis. Use of queuing theory illustrates LWM’s ability to counter insider-threats. Erdos’s random graph theory is then used to prove LWM’s ability to assure the ledger’s global consistency. The talk should provide attendees with an understanding of current DLTs and how they work. Attendees should also get an understanding of how mathematical tools can be used to predict the behavior of complex overlay networks embedded in the global Internet.

Bio: R.R. Brooks is Professor of Electrical and Computer Engineering at Clemson University in Clemson, South Carolina. He received a PhD in Computer Science from Louisiana State University and a B.A. in Mathematical Sciences from The Johns Hopkins University. Dr. Brooks also studied Operations Research at the Conservatoire National des arts et Metiers in Paris, France. He is a senior member of the IEEE and a distinguished member of the ACM. He wrote the books Disruptive Security Technologies with Mobile Code and Peer-to-Peer Networks , Distributed Denial of Service Attacks, and Introduction to Computer and Network Security: Navigating Shades of Gray and co-wrote Multi-Sensor Fusion. He co-edited both versions of Distributed Sensor Networks (with S. S. Iyengar). Dr. Brooks' security research funded by the US Department of State developed secure communications tools being used by activists and journalists avoiding repression by authoritarian regimes. He created tools for both exploiting and foiling side-channel attacks. His team has performed professional penetration testing for clients. The security research of his group is particularly tailored towards networks of embedded systems. Dr. Brooks was PI of the Reactive Sensor Networks Project sponsored by the DARPA ITO Sensor Information Technology initiative, which explored collaborative signal processing to aggregate information moving through the network, and the use of mobile code for coordination among intelligent sensor nodes. Dr. Brooks was co-PI of a DARPA IXO JFACC program that used distributed discrete event controllers for air combat C2 planning. He coordinated a DARPA MURI program that uses cooperating automata in a cellular space to coordinate sensor network planning and execution. Dr. Brooks was PI of an ONR URI on cybersecurity issues relating to mobile code and the construction of secure information infrastructures. Dr. Brooks' research on computer and network security has been sponsored by ONR, DARPA, ARO, AFOSR, NIST, US Department of State, NSF, the US Department of State and BMW Manufacturing Corporation. His Ph. D. dissertation received an exemplary achievement certificate from the Louisiana State University graduate school. Dr. Brooks is Associate Editor of Elsevier computers and Security. He has a broad professional background with computer systems and networks. Dr. Brooks was head of the Pennsylvania State University Applied Research Laboratory Distributed Systems Department for over six years. He was technical director of Radio Free Europe's computer network for many years. His consulting clients include the French stock exchange authority and the World Bank.




April
4

cs Computer Science Department Colloquium
Immersive Analytics: The Conflict In The Land Between the Gulfs.

Speaker: Francisco Ortega, Assistant Professor, Department of Computer Science, Colorado State University.

When: 11:00AM ~ 11:50AM, Monday April 4, 2022
Where: CSB 130 map

Abstract: Within research on 3-dimensional information visualization shown in 3-dimensionally rendered environments (stereoscopic immersive analytics, or just IA in this talk) there are countless open questions. How do we remove the gulf of execution (e.g., what inputs provide the most seamless interactions in IA)? How do we remove the gulf of evaluation (e.g., what visualization methods allow for easy interpretation)? What type of devices should display the content (e.g., is virtual reality (VR) better than augmented reality (AR) for IA)? More succinctly "what combination of display technologies, interaction techniques, and visualization design is optimal?"

The answer is that nobody knows. Natural interaction techniques and modalities can improve user experience and efficiency while navigating IA environments. Even so, most work in IA has been done using a ray-cast or similar pointer-based interaction, as opposed to mid-air gestures or speech commands. Perhaps this is because interactions that work well in one display may not transfer too well when used in another. Consider the difference between using hand-based interactions in VR compared to AR.

If we hold interaction techniques constant, does the rendering technology impact the users' understanding of the environment? AR devices very typically have a narrower field of view and more muted colors compared to VR devices. These differences may cause some IA tasks to become more difficult when done in AR. In AR the user can see their real-world but is seeing one’s real-world blessing or a burden when tasked with IA data exploration. Most would agree that co-located collaboration is improved by awareness of others' non-verbal communications, these communications are mostly preserved when using AR where many non-verbal cues are lost when using VR. That said, VR can provide an environment free of the distractions found in the real world.

This talk is intended to spark a conversation around these and other open areas of IA research.

Bio: Francisco R. Ortega is an Assistant Professor at Colorado State University (CSU) and Director of the Natural User Interaction lab (NUILAB). Dr. Ortega earned his Ph.D. in Computer Science (CS) in the field of Human-Computer Interaction (HCI) and 3D User Interfaces (3DUI) from Florida International University (FIU). He also held the position of Post-Doc and Visiting Assistant Professor at Florida International University between February 2015 to July 2018. Broadly speaking, his research has focused on multimodal and unimodal interaction (gesture-centric), which includes gesture recognition and elicitation (e.g., a form of participatory design). His main research area focuses on improving user interaction by (a) multimodal elicitation, (b) developing interactive techniques, and (c) improving augmented reality visualization techniques. The primary domains for interaction include immersive analytics, assembly, Navy use cases, and collaborative environments using augmented reality headsets. His research has resulted in over 76 peer-reviewed publications including books, journals, conferences, workshops, and magazine articles, among others, in venues such as IEEE TVCG, ACM PACMHCI, ACM ISS, ACM SUI, and IEEE 3DUI, among others. He is the first author of Interaction Design for 3D User Interfaces: The World of Modern Input Devices for Research, Applications, and Game Development book by CRC Press. Dr. Ortega has experience with multiple projects awarded by the government. For example, Dr. Ortega was a co-PI for the DARPA Communicating with Computers project. He is currently a PI for a 3-year effort for ONR titled Perceptual ⁄ Cognitive Aspects of Augmented Reality: Experimental Research and a Computational Model. He has also been funded by the National Science Foundation and sub awardee for an ONR project from Virtual Reality Rehab. Since his initial tenure-track appointment at CSU from August 2018 to January 2022, Dr. Ortega has brought over 2.5 million dollars in external funding, with 1.9 million as principal investigator. Finally, Dr. Ortega is committed to diversity and inclusion and it is his mission of him to increase the number of underrepresented minorities in CS.




April
11

cs Computer Science Department Colloquium
Embodied Computational Metacognition

Speaker: Nikhil Krishnaswamy, Assistant Professor, Department of Computer Science, Colorado State University

When: 11:00AM ~ 11:50AM, Monday April 11, 2022
Where: CSB 130 map

Abstract: Metacognition is defined as an actor's awareness and understanding of its own thought processes and the patterns underlying them. More simply it may be called "thinking about thinking." A metacognitive agent is one that analyzes its own model of the environment and world, including the individual concepts it considers therein, figures out when it needs to be updated, how, and why. While modern AI has demonstrated success at many tasks, the state of the art methods (usually large neural networks) do not easily or organically adapt to accommodate new concepts. In this talk, I discuss how three different AI techniques—machine learning, symbolic reasoning, and embodied simulation—come together to enable rapid bootstrapping of methods for novel concept detection, physical reasoning about objects and events, and language grounding, with fewer resources than typically required by state of the art methods in natural language understanding and multimodal reasoning.

First I will introduce methods for creating virtual and mixed-reality environments and agents to operate within them. Second, I will discuss how one such agent explores its own environment, using a simple example of object reasoning inspired by infant and toddler learning, and demonstrate how embodied simulation, reinforcement learning, and high-dimensional embedding spaces can be exploited to rapidly detect changes in the environment that require the agent's underlying model to be updated. Subsequently, I will discuss how the resulting vector representations can be used to explore where in a neural network abstract properties pertaining to objects, events, and affordances "reside," can be isolated, and how they can be used compositionally. Finally I will demonstrate how representations from two entirely distinct models can be correlated with each other, such as a word with an object, using an example of assigning linguistic labels to acquired concepts, and determining when words are used in different senses, grounded to the environment.

Bio: Nikhil Krishnawamy is an Assistant Professor of Computer Science at Colorado State. His research straddles the boundaries of Artificial Intelligence, Natural Language Processing, Cognitive Science, and Human-Computer Interaction. He earned his Ph.D. at Brandeis University in 2017, followed by a 3-year postdoc funded by the DARPA Communicating with Computers (CwC) program. His main research interests are: using computational methods to examine human cognition and linguistic processing, including statistical, symbolic, neural, and simulation methods; and examining the properties of high-dimensional embedding spaces for representation. His research has been funded by DARPA and NSF, and has appeared at top AI and NLP conferences, including AAAI and ACL venues. He won a best demo award at ICAT-EGVE 2020, the merger of two of the premier conferences in virtual environments, and has served on the program committee for top conferences in AI, NLP, and Cognitive Science, including AAAI, AACL, ACL, CogSci, COLING, EACL, EMNLP, IWCS, NAACL-HLT, and more.




April
18

cs Computer Science Department Colloquium
Detecting and Profiling Visual Disinformation on the Internet

Speaker: Aparna Bharati, Department of Computer Science & Engineering, P. C. Rossin College of Engineering & Applied Sciences, Lehigh University

When: 11:00AM ~ 11:50AM, Monday April 18, 2022
Where: CSB 130 map

Abstract: Visual disinformation has become an important area of research due to the exponential increase in availability and free exchange of media. The increase in sharing of images has also led to more ways to edit image content aimed towards achieving a certain goal. It can be altering one’s portrait to look younger, increasing color contrast in natural scenes, or adding external objects to the images to change their perception and understanding. Edits or manipulations with malicious intent can be used to mislead readers or followers. Images manipulated with benign intent behind them may also be fatal to the society by shifting what is perceived as normal and giving the viewers unrealistic expectations. In order to assess and regulate the quality of media, it is important to devise algorithms that detect and analyze manipulated content in an automated way. My work, in this domain, focuses on solving two such problems in the space of visual mis ⁄ disinformation - detecting instances of manipulation in images and analyzing the provenance of a given image. In addition to understanding single image properties, provenance analysis considers a collection of active variants of the image in question and requires retrieving the stages of evolution of the manipulated media object and the other objects contributing parts to the stages. The talk will present the problem definition and solutions that are applicable to general cases of manipulated images at a large scale. They are evaluated on unconstrained scenarios and tested with large scale datasets. Our proposed pipelines utilize techniques from computer vision and machine learning to solve these important problems in the domain of image forensics.

Bio: Aparna Bharati is an Assistant Professor in the Department of Computer Science & Engineering at Lehigh University. Her research interests include Media Forensics, Computer Vision, Biometrics, Pattern Recognition and Machine Learning. She serves as a member of the IEEE Information Forensics and Security Technical Committee and IEEE Biometric Council's Editorial Board. She has served in the program committee of venues such as ICPR, FG and CVPR workshops. She graduated with her Ph.D. in 2020 from the University of Notre Dame where she was a research assistant for DARPA's MediFor project led by Drs. Kevin Bowyer, Walter Scheirer and Patrick Flynn and a member of the Computer Vision Research Lab. She was a Machine Learning Research Intern in the Document Intelligence Lab at Adobe Research where she worked on document change analysis with Dr. Rajiv Jain and Vlad Morariu. Prior to coming to the US, she graduated from IIIT-Delhi with a B. Tech in Computer Science & Engineering and specialized in Image Analysis and Machine Intelligence. Her undergraduate research was supervised by Dr. Mayank Vatsa and Dr. Richa Singh.




April
25

cs Computer Science Department Colloquium and ISTec Distinguished Lecture Series
Safe and Secure Cyber–Physical and Internet–of–Things (IoT) Systems

Speaker: Dr. Marilyn Wolf, Elmer E. Koch Professor of Engineering School of Computing, University of Nebraska–Lincoln

When: 11:00AM ~ 11:50AM, Monday April 25, 2022
Where: CSB 130 map

Abstract: Cyber–physical systems (CPS) and Internet–of–Things (IoT) systems connect computers to the physical world. CPS and IoT systems are at the heart of our civilization: infrastructure, logistics, medicine. Both the physical safety and information security of these systems is required for continued functioning of these critical systems. Safety and security, while traditionally separate disciplines, are intertwined in CPS and IoT systems. Information security measures are inadequate to properly protect these systems. This talk, based on work with Dimitrios Serpanos, outlines safety and security issues and proposes some approaches.

Bio: Marilyn Wolf is Elmer E. Koch Professor of Engineering and Director of the School of Computing at the University of Nebraska–Lincoln. She received her BS, MS, and PhD in electrical engineering from Stanford University in 1980, 1981, and 1984, respectively. She was with AT&T Bell Laboratories from 1984 to 1989. She was on the faculty of Princeton University from 1989 to 2007 and was Farmer Distinguished Chair at Georgia Tech from 2007 to 2019. Her research interests included embedded computing, embedded video and computer vision, and VLSI systems. She has received the IEEE Kirchmayer Graduate Teaching Award, the IEEE Computer Society Goode Memorial Award, the ASEE Terman Award and IEEE Circuits and Systems Society Education Award. She is a Fellow of the IEEE and ACM and an IEEE Computer Society Golden Core member.

To arrange a meeting with the speaker, please contact Mahdi Nikdast (Mahdi.Nikdast@colostate.edu).




May
2

cs Computer Science Department Colloquium
AI and the Abugida: Automated Transcription of Historical Documents

Speaker: Samuel Grieggs, PhD candidate at the University of Notre Dame.

When: 11:00AM ~ 11:50AM, Monday May 2, 2022
Where: CSB 130 map

Abstract: Researchers in the humanities can spend years going to collections all throughout the world to find the primary sources that will give them the key discoveries that help us understand our cultural heritage. Unfortunately, the ability to access these documents can be limited by all sorts of external factors. From protective archivists, international travel restrictions, and even simple lack of resources, it is not always possible to physically get to them.Therefore, recent efforts have been made to digitize large collections of historical handwritten manuscripts, and make the scanned images available online. The transcription of handwritten historical documents into machine-encoded text has always been a difficult and time-consuming task, and in fact entire academic careers are built around transcribing individual codices and producing a definitive edition. The automatic transcription of handwritten text is known as Handwritten Text Recognition, and it is a robust research area for both modern and historical documents, but there are unique challenges that come when working with historical documents.

Bio: Samuel Grieggs is a PhD candidate at the University of Notre Dame. He is a member of the Computer Vision Research Lab (CVRL), advised by Dr. Walter Scheirer. He earned his B.S. in Computer Science at Indiana University of Pennsylvania in 2017, and M.S. in Computer Science and Engineering at the University of Notre Dame in 2021. Additionally, He was a research intern at Air Force Research Lab in Rome NY in 2015 and 2019. His research interests include: Digital Humanities, Automated Document Analysis, Open Set Recognition, Human Activity Recognition, and Novelty Detection.