Videos in category Computer Science

Computer document processing often starts with an abstract, structural, representation before entering a processing pipeline which creates a desired layout and appearance. But unfortunately the whole system resembles a series of steps in a one-way chemical reaction, or the successive irreversible stages of creating assembler code using a compiler. This `one-way function' behaviour is most obvio...

Fractals and splines have very different geometric features. Fractals can be continuous everywhere, yet differentiable nowhere. Fractals are often selfsimilar curves with fractional dimension. And fractals are also attractors, fixed points of iterated function systems. In contrast, splines are piecewise polynomial curves, so well behaved that they are often used for large scale industrial desig...

This talk describes three of the most beautiful pieces of code that I have ever written: three different implementations of Hoare's classic Quicksort algorithm. The first implementation is a bare-bones function in about a dozen lines of C. The second implementation starts by instrumenting the first program to measure its run time; a dozen systematic code transformations proceed to make it more ...

Text content can allow unintended inferences. Consider, for example, the numerous people who have published anonymous blogs for venting about their employer only to be identified through seemingly non-identifying posts. Similarly, the US government's "Operation Iraqi Freedom Portal" was assembled as evidence of nuclear weapons presence in Iraq, but removed because it could be used to infer much...

Summary Sir Tony Hoare discusses the relationship between the science of computing and the engineering of software. First, he looks at the general concerns of science (e.g. long term, ideal, formality) and engineering (e.g. short-term, adequacy, and dependability) before delving on the interdependencies of the two viewpoints. He concludes with a vision of what software will be like: "Someday .....

I built a balancing robot that uses vision to dance with people using the Parallax Propeller multi-core processor. My talk will cover the robot and the tool I wrote to build and debug sophisticated embedded systems. ViewPort now supports computer vision using OpenCV, fuzzy logic, physics simulation, and monitoring telemetry with simulated instruments. I use Google Checkout, Analytics, CustomSea...

Active learning considers the problem of actively choosing the training data. This is particularly useful in settings where the training data is limited or comes with a price and therefore the learner needs to be "economical" in its data usage. Active learning can be particularly challenging in settings where the cost of the data varies, the learner only has partial control over the data it rec...

Author: David Richards

I give an overview of our current and future work on route planning. Based on contraction hierarchies, a simple technique that allows fast routing by exploiting the hierarchy available in the network, I explain how this can be used for static routing in continent sized road networks in about 0.1 ms, as a basis for transit-node routing that is another two orders of magnitude faster, and for comp...

This talk focuses on some of the key projects that Microsoft Research is developing in their research labs, worldwide. The key project discussion will be supplemented by a number of prototype demos from the 55 areas of computer science research which our labs are engaged in. Behrooz Chitsaz Microsoft Bryan Mistele Inrix