All colloquia take place on Friday from 12:00 to 2:00 pm (unless otherwise noted) in Room D428 and D430 on the fourth floor of the Muenzinger Psychology Building.
If you have questions, please call the ICS Office at 303-492-5063.
January 16, 2017
Martin Luther King Day - University Closed
January 17, 2017
First Day of Classes
January 27, 2017
Assistant Professor, Computer Science, University of Colorado Boulder
Title: The Ground Truth About Metadata and Community Detection in Networks
Abstract: Community detection is one of the most common tasks in network analysis, in which we seek to decompose a network into its underlying structural modules or groups by examining only the pattern of connections in the network. The quality of community detection methods are typically evaluated by how closely the communities they find correlate with node "metadata", which are empirically observed labels on nodes, e.g., a person's ethnicity in a social network or the brain region in a connectome.
In this talk, I will present two results on community detection and node metadata. First, I'll show that it is theoretically impossible, via a No Free Lunch theorem, for one community detection algorithm to be universally better than any other. This result further implies that no community detection method can always recover the "ground truth" communities in a network. However, by using node metadata to guide the community detection process, rather than as an evaluation target, better community detection results can often be obtained. To illustrate this point, I'll introduce a Bayesian model that can learn the correlation between node metadata and network communities, if any exists. The learned correlations are interesting in their own right, and allow us to make predictions about the community membership of nodes whose network connections are unknown. After sketching the method, I'll demonstrate it on synthetic networks with known structure, where the method performs better than any algorithm can without metadata, and on real-world networks, large and small, drawn from social, biological, and technological domains. This is joint work with Leto Peel, Daniel B. Larremore, and Mark Newman.
February 3, 2017
Christopher A. Lowry
Associate Professor, Integrative Physiology, University of Colorado Boulder
Title: An Immunization Strategy for Prevention of Stress-Related Psychiatric Disorders
Abstract: Novel prevention and treatment strategies are urgently needed to reduce the burden of stress-related psychiatric disorders, including posttraumatic stress disorder (PTSD) and major depressive disorder (MDD). Both preclinical and clinical studies suggest that inflammation increases vulnerability to development of anxiety and affective disorders. Consequently, immunoregulatory strategies to decrease inflammation have potential for the prevention and treatment of these disorders. Using a murine model of chronic psychosocial stress, the chronic subordinate colony housing (CSC) model, we found immunization with a heat-killed preparation of Mycobacterium vaccae, a bioimmunomodulatory agent previously shown to activate regulatory T cells (Treg) and to increase production of anti-inflammatory cytokines, prevented development of a PTSD-like syndrome. Immunization with M. vaccae antigen induced a more proactive emotional coping style during exposure to a dominant aggressor, and, in association with suppression of proinflammatory cytokine secretion, prevented stress-induced development of spontaneous colitis and aggravation of colitis in a model of inflammatory bowel disease. Analysis suggests that the protective effects of M. vaccae immunization are due to protection from a stress-induced proinflammatory gut microbial community. Consistent with this hypothesis, protective effects of immunization were absent following Treg depletion. These data provide a hypothetical framework for development of novel strategies for prevention of stress-related psychiatric disorders in vulnerable individuals. Clinical studies investigating the microbiome and potential health benefits of treatment with immunoregulatory probiotics in Veterans with PTSD are ongoing.
February 10, 2017
Hatfield Professor of Law, Executive Director of the Silicon Flatirons Center, Faculty Director, Campus Innovation & Entrepreneurship Initiative, and Dean Emeritus (School of Law); University of Colorado Boulder
Topic: Why Every Academic Needs to Be An Innovator and Entrepreneur
Abstract: The changing dynamics affecting higher education call on all faculty, staff, and students to develop innovative and entrepreneurial mindsets because the world around us is changing faster than ever before. This vision is both compatible with and supportive of the traditional commitment to liberal arts education. It requires, however, that those involved in higher education recognize the changing financial pressures, the evolving employment marketplace, and the competencies that are valued in the 21st century economy and society.
For higher education to adjust and thrive in the twenty first century, it must adopt an innovation mindset in order to ensure that students will benefit from their investment in undergraduate or graduate experiences. Using the challenges faced by law schools as a case study (applications to law school declined by 40% in the wake of the Great Recession), Professor Weiser will explain how an innovation mindset can help the University of Colorado Boulder thrive as the higher education environment continues to evolve.
Title: Streamlining Clinical Science with Structured Data Archives: Data-Driven Insights from the Stroke Rehabilitation Literature.
Abstract: Significant recent advances in bibliometrics have focused on how to utilize text mining approaches to organize a scientific discipline based on author networks, keywords and/or references cited. While these approaches can provide very useful insights, they fail to capture important experimental data that are embedded within many scientific disciplines. A major objective of my work is to examine how experimental data can be used to organize the literature within a discipline and identify its key gaps. This approach is especially important to disciplines that rely heavily on randomized control trials (RCTs), as many of these studies have similar information architecture with common data elements (CDEs) acquired at similar times. Using stroke rehabilitation as an informative example, I will present a massively systematic review of the literature: the Centralized Open-Access Rehabilitation database for Stroke (SCOAR). Using SCOAR as an example, I will show how primary research can be supplemented by meta-scientific research, informatics, and interactive data-visualization. These tools highlight sources of systematic variability in the outcomes of medical interventions and, in turn, provide insights to practitioners and identify opportunities for future research. I hope to make this talk exciting for a broad audience as the meta-scientific approach is fruitful for anyone pursuing experimental work, and medical informatics draws on numerous ICS disciplines from information science (for ontology), to neuroscience (for the subject matter), to computer science (for our interactive visualizations).
March 10, 2017
Assistant Professor, Molecular, Cellular, and Integrative Neurosciences Program, Colorado State University; Director, Sensorimotor Neuroimaging Laboratory
March 23, 2017 *Thursday*
Associate Professor of Electrical and Computer Engineering, University of Wisconsin-Madison
March 27-1, 2017
April 7, 2017
Anna Spain Bradley
Associate Professor of Law; Assistant Vice Provost for Faculty Affairs and Diversity; University of Colorado Law School
April 21, 2017
Associate Professor, Institute of Cognitive Science, University of Colorado Boulder
April 25, 2017 *at 3:00 PM, Tuesday*
Professor of Logic and Metaphysics, University of Edinburgh
May 5, 2017
ICS Poster Session and Mexican Fiesta
Last Day of Classes