Oblivious Sketching for Logistic Regression

               Alexander Munteanu 1 Simon Omlor 2 David P. Woodruff 3

            Abstract                dimensionality reduction technique to compress the data
   What guarantees are possible for solving logistic     to a significantly smaller sketch of polylogarithmic size.
   regression in one pass over a data stream? To       In a second step we feed the sketch to a standard solver
   answer this question, we present the first data      for the  ...