Explore >> Select a destination
|
You are here 
|
talyarkoni.org | ||
| | | | |
www.frontiersin.org
|
|
| | | | | 1 IntroductionThe ego depletion effect has been one of the most cited psychological phenomena since Baumeister et al. first introduced the term in 1998. The ... | |
| | | | |
pedermisager.org
|
|
| | | | | Personal website of Dr. Peder M. Isager | |
| | | | |
jaydaigle.net
|
|
| | | | | This is the second-part of a three-part series on hypothesis testing. Today we'll look at the way we do hypothesis testing in practice, and how it tends to fail. Modern researchers use hypothesis testing as a tool to develop knowledge, but it's really a tool for making decisions, and so it encourages us to draw strong conclusions from weak evidence. It also encourages us to view studies that don't reject the null hypothesis as failures, which leads even honest and dedicated researchers to do shoddy resea... | |
| | | | |
iclr-blogposts.github.io
|
|
| | | The product between the Hessian of a function and a vector, the Hessian-vector product (HVP), is a fundamental quantity to study the variation of a function. It is ubiquitous in traditional optimization and machine learning. However, the computation of HVPs is often considered prohibitive in the context of deep learning, driving practitioners to use proxy quantities to evaluate the loss geometry. Standard automatic differentiation theory predicts that the computational complexity of an HVP is of the same order of magnitude as the complexity of computing a gradient. The goal of this blog post is to provide a practical counterpart to this theoretical result, showing that modern automatic differentiation frameworks, JAX and PyTorch, allow for efficient computat... | ||
