Algorithmic Aspects of Machine Learning

Ankur Moitra wrote:In this book, we will approach the problem of giving provable guarantees for machine learning by trying to find more realistic models for our data. In many applications, there are reasonable assumptions we can make based on the context in which the problem came up, that can get us around these worst-case impediments and allow us to rigorously analyze heuristics that are used in practice, as well as design fundamentally new ways of solving some of the central, recurring problems in machine learning.