Currently set to No Index

Brand Safety Requires Effective Identification Measures

White Paper from Advertising Standards Organizations Urges Industry to Adopt Identifiers Across Ad Assets, Businesses, and Consumers
A new white paper released by five leading advertising standards organizations highlights the importance for brand safety of adopting effective identification measures across the digital advertising supply chain.

Read Full Story

Finding out why

Paper: Compositional Hierarchical Tensor Factorization: Representing Hierarchical Intrinsic and Extrinsic Causal Factors
Visual objects are composed of a recursive hierarchy of perceptual wholes and parts, whose properties, such as shape, reflectance, and color, constitute a hierarchy of intrinsic causal factors of object appearance.

Read Full Story

Power and Sample Size for Repeated Measures ANOVA with R

Background
One of my colleagues is an academic physical therapist (PT), and he’s working on a paper to his colleagues related to power, sample size, and navigating the thicket of trouble that surrounds those two things. We recently got together to walk through some of the issues, and I thought I would share some of the wildlife we observed along the way.

Read Full Story

Why Artificial Intelligence Research is Still Relevant

As a college student, you will often be required to submit a written research paper in whatever field. For such a paper, you will have to analyze different data and come up with a hypothesis. This is where artificial intelligence comes into play. It helps you analyze data and make predictions based on your findings.

Read Full Story

Truncated Bi-Level Optimization

In 2012, I wrote a paper that I probably should have called “truncated bi-level optimization”.  I vaguely remembered telling the reviewers I would release some code, so I’m finally getting around to it.
The idea of bilevel optimization is quite simple.  Imagine that you would like to minimize some function .  However, itself is defined through some optimization.

Read Full Story

Hyper Networks | 大トロ

In this post, I will talk about our recent paper called [1609.09106] HyperNetworks. I worked on this paper as a Google Brain Resident – a great research program where we can work on machine learning research for a whole year, with a salary and benefits! The Brain team is now accepting applications for the 2017 program: see g.co/brainresidency.

Read Full Story

Some misc news

I just learned my postdoc roommate Yisong Yue from Caltech released a new interesting paper: Factorized Variational Autoencoders for Modeling Audience Reactions to Movies: a joint work with Disney Research, published @ CVPR 2017. Another interesting paper: Accelerating Innovation Through Analogy Mining, just received the best paper award at KDD 2017.

Read Full Story

Where Predictive Modeling Goes Astray

I recently reread Yarkoni and Westfall’s in-progress paper, “Choosing prediction over explanation in psychology: Lessons from machine learning”. I like this paper even more than I liked their previous paper, but I think a few notes of caution should be raised about the ways in which a transition to predictive modeling could create problems that psychologists are seldom trained to deal with.

Read Full Story

On Marginal Likelihood and Cross-Validation

Here’s a paper someone has pointed me to, along the lines of “everything that works, works because it’s Bayesian”:Edwin Fong, Chris Holmes (2019) On the marginal likelihood and cross-validationI found this paper to be lacking on the accessibility front, mostly owing to the fact that it is a mixture of two somewhat related but separate things:(A) a simple-in-hindsight and c

Read Full Story

Notes on the Limitations of the Empirical Fisher Approximation

This post is a short not on an excellent recent paper on empirical Fisher information matrices:Kunstner, Balles and Hennig (2019) Limitations of the Empirical Fisher ApproximationI was debating with myself whether I should write a post about this because it’s a superbly written paper that you should probably read in full.

Read Full Story

Twitter, Social Bots, and the US Presidential Elections!

Our paper titled Social bots distort the 2016 U.S. Presidential election online discussion was published on the November 2016 issue of First Monday and selected as Editor’s featured article!
We investigated how social bots, automatic accounts that populate the Twitter-sphere, are distorting the online discussion about the 2016 U.S. Presidential elections.

Read Full Story

Why Responsible AI Development Needs Cooperation on Safety

We’ve written a policy research paper identifying four strategies that can be used today to improve the likelihood of long-term industry cooperation on safety norms in AI: communicating risks and benefits, technical collaboration, increased transparency, and incentivizing standards.

Read Full Story