IHP just organized a series of talks within what they call the nexus trimester whuch had a focus on Inference Problems -the whole playlist is here-. Here are a few presentations related to the theme of Nuit Blanche.
Sketching and Embeddings , Alex Andoni at IHP
Abstract: Sketching for distance estimation is the problem where we need to design a possibly randomized function f from a metric space to short strings, such that from f(x) and f(y) we can estimate the distance between x and y. This problem is a core problem in both the streaming and nearest neighbor search areas. We will discuss this problem and its connections to the theory of metric embeddings. In particular, we will discuss when and why sketching is equivalent to embedding into normed space such as ℓ1.
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.