Class Type: Summer 2016
Regularizer design for structured learning
Design and analysis of tractable methods for estimation of structured models from massive high-dimensional datasets has been a topic of research in statistics, machine learning and engineering for many years. Regularization, the idea of simultaneously optimizing a data fidelity term (loss) and a structure-promoting term (regularizer), is a widely used …
The XyloTron: A machine vision-based wood identification device
The worldwide value of illegal logging is 30-100 billion USD annually and within a number of key producer countries an estimated 50-90% of the forestry is illegal. Approximately 80% of the worldwide illegal logging is controlled by organized crime. Rapid, accurate, and inexpensive wood identification is an important factor in …
A big conjecture in combinatorics that unexpectedly got proved last week and what this has to do with matrix multiplication
There’s a popular folk question about the card game Set: how many cards can you have on the table before there’s necessarily a legal play? The answer, which is pretty tricky to work out, and which is larger than most people expect, is 20. This is one case of a …