Feb 1, 2012. Eli Cortez
Title: Information Extraction over Textual Sources
Abstract: The growing use of text files for information exchange, such as HTML pages, XML documents, e-mail, blogs posts, tweets, RSS and SMS messages, brings numerous problems related to how to properly exploit the information contained therein. In particular, problems related to Information Extraction (IE) from text have motivated several works in various scientific communities in areas such as Databases, Data Mining, Information Retrieval and Artificial Intelligence. In this talk, it will be presented an overview of the IE problem and methods that have been proposed in recent literature to deal with it. The IE problem consists in extracting values of interest arranged in unstructured texts, such as postal addresses, bibliographic citations, classified ads, that are implicitly present in textual sources from a variety of different domains. It will be discussed the main and most recent approaches proposed in the literature, with particular emphasis on probabilistic methods.
Feb 8, 2012. Chris Bregler
Feb 15, 2012. Jonathan Viventi
Title: High-Resolution Brain Machine Interfaces using Flexible SiliconElectronics: Research and Clinical Applications
Current implantable brain devices for clinical and research applications require that each electrode is individually wired to a separate electronic system. Establishing a high-resolution interface over broad regions of the brain is infeasible under this constraint, as an electrode array with thousands of passive contacts would require thousands of wires to be individually connected. To overcome this limitation, we have developed new implantable electrode array technology that incorporates active, flexible electronics. This technology has enabled extremely flexible arrays of 720 and soon, thousands of multiplexed and amplified sensors spaced as closely as 250 µm apart, which are connected using just a few wires. These devices yield an unprecedented level of spatial and temporal micro-electrocorticographic (µECoG) resolution for recording and stimulating distributed neural networks. µECoG is one of the many possible applications of this technology, which also include cardiac, peripheral nerve and retinal prosthetic devices. I will present the development of this technology and examples of retinotopic and tonotopic maps produced from in vivo recordings. I will also present examples of finely detailed spatial and temporal patterns from feline neocortex that give rise to seizures and suggest new stimulation paradigms to treat epilepsy.
Feb 22, 2012. Marcel Hlawatsch
Mar 14, 2012. Richard Bonneau
Title: Overview of Richard Bonneau's Research
Our lab is focused on a number of computational biology problems that, if solved, would remove key bottlenecks in biology and systems biology. We focus on two main categories of computational biology: learning networks from functional genomics data and predicting and modeling protein structure. In the area of structure prediction we were early contributors to the Rosetta code; a platform for structure prediction, design and docking. In the area of network inference we worked on two computational methods that were used to demonstrate the first predictive genome-wide model of regulatory dynamics (i.e. the first case where a genome-wide model could predict the whole transcriptional state of cells at future time points not part of the training set). Both network inference and protein structure prediction remain grand challenges and in spite of our progress much exciting work remains to be done in the coming years as we continue to improve, scale and apply these methods.