Events

Printer-friendly version

Talks and Lectures:

Invited Guests and Lecturers | Cognitive Science Colloquia | Potential Events
Current Events | Upcoming Events | Future Events | Recent Events | Past Events

Current Events

Ontology Engineering Spring 2017
Description:
To learn how to build computer understandable definitions of terms for usage in automated systems.
This course provides an introduction to ontologies, their uses, and an overview of their application in semantically enabled systems. Ontologies encode term meanings. Ontologies with their declarative encodings of meaning can be used to improve communications between people and can enable computer programs to function more effectively. They provide the foundation for clear and unambiguous interaction. Ontologies have become increasingly common on the web, and class participants will not only learn about the use of ontologies in web-based applications but how to evaluate ontologies for reuse in such applications. Participants will read relevant papers, learn how to critically review ontology papers as well as ontologies themselves, and will participate in at least one group project designing, using, and evaluating ontologies.

Dates: January 1, 2017 - May 31, 2017
Concepts: Rule Modeling, Semantic Reasoning, Information Model, Linked Data, Taxonomy, Controlled Vocabulary, , Ontology, Semantic Web Services, Semantic Web, Inference, Vocabulary, Schema, Provenance
Data Analytics 2017
Description:
Data and Information analytics extends analysis (descriptive and predictive models to obtain knowledge from data) by using insight from analyses to recommend action or to guide and communicate decision-making. Thus, analytics is not so much concerned with individual analyses or analysis steps, but with an entire methodology. The world at-large is confronted with increasingly larger and complex sets of structured/unstructured information; from sensors, instruments, and generated by computer simulations; data is "hidden" in websites, application servers, social networks and on mobile devices. As a nation, assimilating information across disparate domains (e.g., intelligence, economics, science) has the potential to provide improved capabilities for decision makers. In commerce and industry, analytics-driven enterprises are becoming mainstream. Yet, there is a shortfall in the key education skills needed to meet the growing needs. Traditional enterprises are moving toward analytics-driven approaches for core business functions. In the government and corporations, cybersecurity problems are prevalent. The investment in advanced analytics capabilities could potentially be more broadly leveraged today and greater than any prior government investments in computing. Emphasis is now placed on disruptive data and information sources on the Web and Internet: using Web Science and informatics to explore social networks, platform competition, the "long tail" and economic or resource impacts of the search for new findings. Key topics include: advanced statistical computing theory, multivariate analysis, and application of computer science courses such as data mining and machine learning and change detection by uncovering unexpected patterns in data.
  • Introduce students to relevant methods to recognize and apply quantitative algorithms, techniques and interpretation
  • To develop students' strategic thinking skills, combined with a solid technical foundation in data and model-driven decision-making.
  • Develop ability to apply critical and analytical methods to formulate and solve science, engineering, medical, and business problems
  • Students will examine real-world examples using modern cyberinfrastructure to place statistical and data-mining techniques in context, to develop data-analytic thinking, and to illustrate that proper application is as much an art as it is a science.
  • By the end of the course, students can effectively communicate analytic findings to non-specialists

Dates: January 16, 2017 - May 5, 2017
Concepts: Predictive Analytics, Big Data, Data Science, Analytics, Data Visualization


Recent Events

TWed Discussion: Fun with GANs
Description:
TWed Talk: Weds, 26 Apr (6p Winslow 1140)
TITLE: "Fun with GANs"
LEADER: Matt Klawonn
VIDEO: TWed video streams
EVENT: YouTube
KEYWORDS: Machine Learning, Computer Vision and Pattern Recognition; Neural and Evolutionary Computing

Please join us next Weds, (26 Apr, 6p Winslow 1140) as TWC Ph.D. student Matt Klawonn leads us in a discussion of an exciting new area of machine learning known as Generative Adversarial Networks (GANs). One well-known application of GANs has been the creation of photorealistic images.

DESCRIPTION: According to Yann LeCun, "There are many interesting recent developments in deep learning ... The most important one, in my opinion, is adversarial training (also called GANs for Generative Adversarial Networks). This, and the variations that are now being proposed is the most interesting idea in the last 10 years in ML, in my opinion." In this talk we will introduce GANs, starting with some theory and moving to implementation tips and techniques. We will then take a look at some GAN demonstrations and code, moving from image to sequence generation.

BIO: Matt Klawonn is a third-year PhD student working under adviser Jim Hendler in various deep learning areas. His most recent project, one which he hopes to turn into a thesis, involves the use of GANs to create knowledge graphs from images.

READINGS:
  1. "An introduction to Generative Adversarial Networks (with code in TensorFlow)."
  2. Goodfellow, Ian J.; Pouget-Abadie, Jean; Mirza, Mehdi; Xu, Bing; Warde-Farley, David; Ozair, Sherjil; Courville, Aaron; Bengio, Yoshua (2014). "Generative Adversarial Networks".
  3. Salimans, Tim; Goodfellow, Ian; Zaremba, Wojciech; Cheung, Vicki; Radford, Alec; Chen, Xi (2016). "Improved Techniques for Training GANs"
REMINDERS:
  • ALL TWC STUDENTS are STRONGLY encouraged to attend, regardless of whether the talk is in your specific research area.
  • SIGN UP NOW for your Spring 2017 LIGHTNING TALK!
  • Sign up for RPIrates, the RPI R Users Group
===
TWed Logistics (Spring 2017):

Dates: April 26, 2017 - April 26, 2017
Concepts: Computer Vision, Neural Computation, Pattern Recognition, Evolutionary Computation, Machine Learning
"A New Vision for Dataset Versioning
Description:
TWed Talk: Weds, 19 Apr (6p Winslow 1140)
TITLE: "A New Vision for Dataset Versioning" (Tentative Title)
LEADER: Benno Lee
VIDEO: TWed video streams
EVENT: YouTube

Please join us next Weds (19 Apr, 6p Winslow 1140) as TWC Ph.D. student Benno Lee leads us in a discussion of his research exploring novel approaches to representing the versioning of datasets.

DESCRIPTION: With the proliferation of digital data gathering, data sets rarely remain unchanged through their lifespans. Whether due to correcting errors or updated algorithms, data often needs to be republished. This leads to a demand for the means to identify and communicate about different iterations of the same set of data. The current means of this discussion falls under data provenance, but perhaps a more detailed and valuable discussion may be had with the vocabulary and concepts of versioning. This talk will focus on a linked data model which looks at a prospective data relationship, publishing those relationships in a human and machine readable change log, and beginning the discussion of measuring change.

BIO: Benno Lee has been a Ph.D. student with the Tetherless World Constellation for five and a half years.

REMINDERS:
  • ALL TWC STUDENTS are STRONGLY encouraged to attend, regardless of whether the talk is in your specific research area.
  • SIGN UP NOW for your Spring 2017 LIGHTNING TALK!
  • Sign up for RPIrates, the RPI R Users Group here
===
TWed Logistics (Spring 2017):
  • TWed schedule
  • Snacks will be provided for TWed Talks
  • Live video streams of TWed Talks are now available via YouTube

Dates: April 19, 2017 - April 19, 2017
Concepts:
TWed Discussion: SPARQL: Beyond the BGP
Description:
TWed Talk: Weds, 5 Apr (6p Winslow 1140)
TITLE: "SPARQL: Beyond the BGP"
LEADER: Jim McCusker
VIDEO: TWed video streams
EVENT: YouTube
KEYWORDS: Semantic Web, SPARQL, Linked Data, Graph Patterns

Please join us Weds (5 Apr, 6p Winslow 1140) as TWC DirDataOps Jim McCusker leads us in a practical and revealing exploration of the deeper mysteries of SPARQL, the protocol and query language without which our lives would be more difficult!

DESCRIPTION: SPARQL is much more that basic graph patterns. It's the only query language that lets you combine advanced graph traversal, dynamic database federation, dynamic data segmentation (using named graphs), subqueries, data and knowledge introspection, and inefficient filtering all into a single query. Learn how to do these things (and why you might not always want to) at this week's TWed Talk. We will be querying DBpedia and bringing it to it's knees within 20 minutes. We might talk a bit about how to query nanopublications in interesting ways using SPARQL. Learn about compositional design, and why you might want to be careful with that.

NOTE! All TWC grad students and URPs need to learn SPARQL! If you've been told you need to learn SPARQL, the web tutorials aren't enough; this talk is "required reading."

BIO: BIO: Jim McCusker, PhD is a relic of a bygone era, when graphs were still unnamed and grad students learned query languages straight from the W3C specs.

REMINDERS:
  • ALL TWC STUDENTS are STRONGLY encouraged to attend, regardless of whether the talk is in your specific research area.
  • SIGN UP NOW for your Spring 2017 LIGHTNING TALK!
  • Sign up for RPIrates, the RPI R Users Group here
===
TWed Logistics (Spring 2017):

Dates: April 5, 2017 - April 5, 2017
Concepts: Linked Data, Semantic Web
TWed Discussion: Using Semantic Data Dictionaries for Semantic Data Conversion in SETLr
Description:
TWed Talk: Weds, 29 Mar (6p Winslow 1140)
TITLE: "Using Semantic Data Dictionaries for Semantic Data Conversion in SETLr"
LEADER: Katie Chastain
VIDEO: TWed video streams
EVENT: YouTube
KEYWORDS: Linked Data, Semantic Web, SETLr

Please join us this Weds (29 Mar, 6p Winslow 1140) as TWC PhD student Katie Chastain leads us in a tour-de-force discussion of the challenges presented when including data dictionaries and codebooks in knowledge graphs, and the use of SETLr to slay such dragons. Katie's talk will include a very brief review of SETLr for noobies...

DESCRIPTION: Conversion of tabular data into linked data is a daunting task for many projects, where hand-crafting conversion parameters is not a scaleable solution. After a brief (re)introduction to SETLr and its functionality and capabilities, I will present a Semantic Data Dictionary as an extension to a traditional data dictionary. I will highlight its usefulness for limiting the workload for human users in annotating implicit data structure, and then describe a "parsing" script for processing a completed SDD into a SETLr script. I can talk about potential future work for the parse script, and I invite discussion and ideas for what functionalities may be helpful to other research projects.

BIO: Katie Chastain is a Ph.D student working with Prof. McGuinness, focusing on data curation for semantic web applications. Their current research is with the Gates Foundation's Healthy Birth, Growth, and Development program, working on a relatively painless way to get the many datasets involved into linked data format for use in the cool visualizations the TWC is developing with CASE. Outside of research, their interests include '90s science fiction television, Magic the Gathering, and cooperative problem solving in a fantasy environment (aka Dungeons and Dragons).

REMINDERS:
  • ALL TWC STUDENTS are STRONGLY encouraged to attend, regardless of whether the talk is in your specific research area.
  • SIGN UP NOW for your Spring 2017 LIGHTNING TALK!
  • Sign up for RPIrates, the RPI R Users Group here
===
TWed Logistics (Spring 2017):

Dates: March 29, 2017 - March 29, 2017
Concepts: Semantic Web, Linked Data
TWed Discussion: Magellan - An ontology-driven in-browser faceted data explorer
Description:
TWed Talk: Weds, 08 Mar (6p Winslow 1140)
TITLE: "Magellan - An ontology-driven in-browser faceted data explorer"
LEADER: Alexander Schwartzberg
VIDEO: TWed video streams
EVENT: Youtube

Please join us TOMORROW (Weds, 08 Mar, 6p Winslow 1140) as Alex Schwartzberg leads us in a discussion of a new lightweight ontology-driven JSON-LD explorer that has emerged from his recent DARPA-related work.

DESCRIPTION: Magellan is an open-source faceted data explorer that can browse and query arbitrary JSON-LD datasets. The primary objective of the tool is to provide a portable web-based platform for faceted browsing that is configurable via a user interface rather than code. Magellan leverages ontologies referenced by the datasets it consumes to provide a clear semantic layer to its faceted browser. The presentation will be a discussion of the inception, goals, architecture, development, and future of the Magellan data explorer application.

BIO: Alexander Schwartzberg is an undergraduate researcher at Rensselaer Polytechnic Institute. Prior to working with the DARPA Advanced Manufacturing team at Tetherless World, he worked in industry for three years as a web developer at Cisco Systems and Cortical Metrics. When he=E2=80=99s not developing web applications, Alexander pursues hardware design, product development, 3d printing, and jazz guitar.

REMINDERS:
  • ALL TWC STUDENTS are STRONGLY encouraged to attend, regardless of whether the talk is in your specific research area.
  • SIGN UP NOW for your Spring 2017 LIGHTNING TALK!
  • Sign up for RPIrates, the RPI R Users Group
===
TWed Logistics (Spring 2017):

Dates: March 8, 2017 - March 8, 2017
Concepts:
TWed discussion: Constructing and Maintaining CHEAR - A Community-Built and Evolved Ontology
Description:
TWed Talk: Weds, 01 Mar (6p Winslow 1140)
TITLE: "Constructing and Maintaining CHEAR - A Community-Built and Evolved Ontology"
LEADER: Sabbir Rashid
VIDEO: TWed video streams
EVENT: YouTube

Please join us Weds (01 Mar, 6p Winslow 1140) as TWC grad student Sabbir Rashid leads us in a discussion of work related to an upcoming paper, "A Community-Built and Evolved Ontology and Data Standard for Childhood Health." Sabbir will also cover HADatAc and give a quick demo on one or two CHEAR related use cases using the system.

DESCRIPTION: Sabbir will discuss some of the steps involved in the building of the CHEAR Ontology, including foundational supporting ontologies, integration of pilot proposals into the ontology, and Management and Browsing using HADatAc, which he will then demo in the context of Use Cases that depict how an Analyst may use the acquisition framework and ontology.

BIO: After graduating undergrad with a double major in Physics and Electrical Engineering, Sabbir Rashid spent a year teaching high school and another as a Systems Engineer for General Dynamics. He began graduate school in August of 2015, when he entered RPI as a Robotics student. He found himself shifting from a robotics focus to computer vision, and then from machine learning to computer and web science. By the end of the first school year he began a research assistantship under Deborah McGuinness, with whom he is conducting research into Semantic Web Technologies and transferred to the Computer Science department. Specifically, his work related to semantically annotated data, a field which will allows for the increased understanding of relationships between concepts through data. He is funded on the NIH CHEAR project, which attempts to solve important problems related to Childhood Health Exposures. His work focuses on both the data acquisition and annotation sides, which includes creating an ontology of important concepts related to child health studies, as well as tackling problems related to the ingestion and annotation of the data.

REMINDERS:
  • ALL TWC STUDENTS are STRONGLY encouraged to attend, regardless of whether the talk is in your specific research area.
  • SIGN UP NOW for your Spring 2017 LIGHTNING TALK!
  • Sign up for RPIrates, the RPI R Users Group here
===
TWed Logistics (Spring 2017):

Dates: March 1, 2017 - March 1, 2017
Concepts: