New research initiatives

The School's faculty are embarking on many exiting new projects and collaborations. A few are highlighted below.

Professor Rod Grupen's proposal, "Representing and Exploiting Cumulative Experience with Objects for Autonomous Manipulation," was one of five proposals nationwide selected by NASA's (National Aeronautics and Space Administration) Game-Changing Technology Program.  Grupen will work with Associate Professor Erik Learned-Miller on developing robots capable of solving new problems during deep space exploration. Their project is focused on autonomous sensing and control for multi-fingered grasping and manipulation applications, so that robots can learn how to improvise and trouble-shoot in space missions.

With a 4.5 year award from the Air Force Research Laboratory as part of their Deep Exploration and Filtering of Text (DEFT) Program, Professor Andrew McCallum will work on large-scale probabilistic entity resolution with distributed computation and relation extraction with universal schema using tensor factorization. The DEFT program was created to "harness the power of natural language processing" to automate the discovery of relevant information contained within the large amount of data collected. 

Research Associate Professor R. Manmatha received two Mellon Foundation awards this fall. He and collaborators Professor James Allan and Northeastern University Assistant Professor David Smith received a grant for their pilot project, "Proteus Infrastructure: Work Aggregation and Entity Extraction," to build and evaluate research infrastructure for scanned books. While there are several large scanned book collections, much of these are unstructured and not easily used by scholars in the humanities. The grant will support building a Proteus infrastructure that will help scholars navigate and use such collections more easily. Components of the infrastructure include automatically identifying a book's language, linking multiple editions of canonical works, finding quotations in canonical works, and entity detection.

In the other Mellon grant, Manmatha is collaborating with Texas A&M and other institutions on "OCRing Early Modern Texts." As part of this grant, the researchers will recognize the text from the 18th century English books using optical character recognition systems, and they will use their technology to automatically estimate OCR errors and correct the output of multiple OCR engines.

Associate Professor Ramesh Sitaraman, in collaboration with Professor Prashant Shenoy and graduate student Vimal Mathew, is working on an energy-efficient cloud system that doesn't sacrifice performance. Their recent research focuses on algorithms that turn off servers in a Content Delivery Network (CDN) to save energy. An evaluation of these new algorithms shows it can reduce total energy usage of a CDN by almost 50% while maintaining high performance to users and minimizing the wear-and-tear of the server hardware. Full details on the cloud system at www.umass.edu/researchnext/internet-goes-green.