College of Computing and Informatics
1:00 - 2:00pm | Woodward 106
Speaker: Dr. Yuanyuan Yang
IEEE Fellow; Program Director, US National Science Foundation; SUNY Distinguished Professor, Stony Brook University,
Title: A Vision towards Pervasive Edge Computing
Abstract: This talk presents an emerging pervasive edge computing paradigm where heterogeneous edge devices (e.g., smartphones, tablets, IoT and vehicles) can collaborate to sense, process data and create many novel applications at network edge. We propose a data centric design where data become self-sufficient entities that are stored, referenced independently from their producers. This enables us to design efficient and robust data discovery, retrieval and caching mechanisms. The future research agenda including scalable data discovery, cache management, autonomous processing, trust, security and privacy, incentives and semantic data naming) will be discussed.
Speaker: Dr. Metin Gurcan
Director, Center for Biomedical Informatics Professor, Department of Internal Medicine Wake Forest School of Medicine
Title: Pixels to Diagnosis: How to design, develop and evaluate a medical image analysis system?
Abstract: Increased interest in medical imaging has resulted in development of a variety of image analysis systems. Many of these systems follow the ‘computer-aided diagnosis’ paradigm. In this paradigm, the main function of the image analysis system is to help medical professionals (e.g. radiologists, pathologists, dermatologists) in their decision-making, instead of making decisions on their behalf. If a system is designed to help medical professionals, its logic, development methodology and evaluation should make sense to the medical professionals who use them.
In this talk, we will describe how to develop an image analysis system, its modular blocks, how to translate medical knowledge into algorithms, how to supplement this knowledge with pattern recognition methods, and how to evaluate them with carefully designed reader studies in which medical professionals with varying levels of experience participate as readers.
Speaker: Dr. Michael Lynch
Director, Center for Mechanisms of Evolution, Arizona State University
Title: Mutation, Drift, and The Origin Of Subcellular Features
Abstract: Although natural selection may be the most powerful force in the biological world, it is not all powerful. As a consequence, many aspects of evolution at the genomic can only be explained by the inability of natural selection to operate. This general principle appears to extend to numerous cellular features: the evolution of the ~1000-fold range in mutation rates that exists among species; greatly elevated rates of transcription error; the divergence of the multimeric states of proteins; the phylogenetic drift of gene-regulatory vocabulary; and the scaling of growth rate with organismal size.
An attempt will be made to describe how two diverse sets of observations – phylogenetic variation in error rates at the level of DNA and RNA, and in the maximum growth potential of species – can be explained at the theoretical level. A fundamental principle is that although natural selection relentlessly pushes traits to the highest possible level of refinement, the limits to perfection are dictated by the power of random genetic drift rather than by intrinsic molecular limitations / cellular constraints. This drift-barrier hypothesis broadly implies that the population-genetic environment imposes a fundamental constraint on the paths that are open vs. closed for evolutionary exploration in different phylogenetic lineages, hence defining the possible patterns of adaptation seen at the molecular and cellular level.
Speaker: Dr. Chengxiang Zhai
Professor, Engineering and Computer Science, University of Illinois Urbana Champaign (UIUC)
Title: TextScope - Enhance Human Perception via Intelligent Text Retrieval and Mining
Abstract: Recent years have seen a dramatic growth of natural language text data, including, e.g., web pages, news articles, scientific literature, emails, enterprise documents, and social media such as blog articles, forum posts, product reviews, and tweets. Text data contain all kinds of knowledge about the world and human opinions and preferences, thus offering great opportunities for mining actionable knowledge from vast amounts of text data ("big text data") to support user tasks and optimize decision making in all application domains. However, computers cannot yet accurately understand unrestricted natural language; as such, how to analyze and mine big text data effectively and efficiently is a difficult challenge, and involving humans in a loop of interactive retrieval and mining of text data is essential.
In this talk, I will present the vision of TextScope, an interactive software tool to enable users to perform intelligent information retrieval and text analysis in a unified task-support framework. Just as a microscope allows us to see things in the ìmicro world,î and a telescope allows us to see things far away, the envisioned TextScope would allow us to ìseeî useful hidden knowledge buried in large amounts of text data that would otherwise be unknown to us. As examples of techniques that can be used to build a TextScope, I will present some of our recent work on general algorithms for analyzing text and non-text data jointly to discover interesting patterns and knowledge. At the end, I will discuss the major challenges in developing a TextScope and some important directions for future research.