On Tuesday, November 1 at noon, Gábor Bojár, President of the Aquincum Institute of Technology (AIT), will give a presentation on their study abroad program in Noyce 3821.
An upcoming talk will be presented on campus to encourage students to consider a great new study abroad program, Aquincum Institute of Technology BUDAPEST, for students interested in computing, design, computational biology, and IT entrepreneurship.
About AIT: The AIT program has a first-rate faculty including professors such as Erno Rubik (inventor of the Rubik's Cube and recent recipient of the U.S. Outstanding Contributions to Science Education Award), an innovative curriculum including courses such as "Computer Vision for Digital Film Post-production" taught by faculty affiliates from Colorfront Studios (recent recipients of an Academy Award for technical contributions), and a guest lecture series that brings prominent speakers to campus.
All classes are conducted in English at AIT's state-of-the-art campus on the lovely banks of the Danube River. Students live in vibrant neighborhoods of Budapest and have ample opportunities to interact with Hungarian students and explore Hungary and the region.
AIT is small and friendly, with typical class sizes of 5-15 students. Recent U.S. AIT students have come from Franklin W. Olin College of Engineering, Harvey Mudd College, Northeastern University, Pomona College, Princeton University, RPI, Skidmore, Smith, Swarthmore and Williams Colleges. The program also includes a small number of Hungarian students. (AIT Alumni).
The AIT website and APPLICATION materials are available on-line.
This Friday at CS Table, Kim and Radhika will be leading a discussion about Online Health Communities. The readings can be found at the following link: http://foswiki.cs.grinnell.edu/foswiki/bin/view/Courses/HealthCareAndComputing/OnlineHealthCommunities
We'll meet at noon in JRC 224A. Please feel free to e-mail [raymondw] if you have any questions!
Developers are increasingly building large software in the form of highly configurable systems, systems with features that can be toggled on and off. The major risk for highly configurable systems is that some bugs, called configuration-dependent faults, only cause failures when certain features are combined, being invisible otherwise. My talk will first discuss the techniques we currently have to combat configuration-dependent faults and show that they all exploit a common idea, which we term feature locality. I will then present some newly discovered forms of feature locality and explain how they are helping us better prevent, find, mitigate, and repair configuration-dependent faults.Refreshments will be served at 4:15 p.m. in the Computer Science Commons (Noyce 3817). Mr. Garvin's talk, "Configuration-dependent faults and feature locality," will follow at 4:30 p.m. in Noyce 3821. Everyone is welcome to attend.
Scientists have been working in the field of text recognition, the science of automatically reading text, for over 200 years. While the problem of reading whole documents (commonly called OCR, or optical character recognition) is more or less solved, the problem of reading text from arbitrary real-world scenes (Scene Text Recognition, or STR) still presents researchers with many challenges. Yet humans have been able to read such text ever since we created language. Many have created a robust recognition programs, but some still suffer from not knowing where the text baseline is—that is, where the non-descending characters of a line of text end. In this talk, we will discuss what makes reading scene text so difficult, how we made a baseline detection algorithm to improve the results of scene text recognition systems, and how we used the scientific method to make our system as robust as possible in ten weeks.Refreshments will be served at 4:15 p.m. in the Computer Science Commons (Noyce 3817). The talk, "A robust system for discovering text baselines in scene text images," will follow at 4:30 p.m. in Noyce 3821. Everyone is welcome to attend.
As the computational power available has grown, the field of machine translation has shifted from using rule-based approaches to statistical-based ones. In essence, many modern machine translation systems learn how to translate by "reading" lots of parallel texts (the same text translated into two languages). The usefulness of this method is largely determined by the amount of parallel texts that are available. This summer I worked at the University of Colorado at Colorado Springs with Dr. Jugal Kalita to create method capable of automatically generating these parallel texts for 92 language pairs.Refreshments will be served at 4:15 p.m. in the Computer Science Commons (Noyce 3817). Mr. Kaufmann's talk, "Automatically generating parallel corpora," will follow at 4:30 p.m. in Noyce 3821. Everyone is welcome to attend.