Regulate This! A New Freakonomics Radio Podcast

Regulate This! A New Freakonomics Radio Podcast

A battle is being waged between the Internet and the State, and this episode of Freakonomics Radio gives you front-row seats. It’s called “Regulate This!” (You can subscribe to the podcast at iTunes, get the RSS feed, or listen via the media player above. You can also read the transcript; it includes credits for the music you’ll hear in the episode.)

At issue is the so-called sharing economy, a range of services that facilitate peer-to-peer transactions through the Internet. Companies like Airbnb, Uber, and Lyft have seen rapid growth and eye-popping valuations, but as they expand around the world, they are increasingly butting heads with government regulators.

In this episode, you’ll hear from Nathan Blecharczyk, the co-founder and CTO of Airbnb (now valued at roughly $10 billion), and one of the youngest billionaires in the world. Blecharczyk tells Stephen Dubner the story of Airbnb’s founding, how it initially struggled to find investors, and what kind of obstacles it still faces daily. In New York City, for instance, it’s estimated that about two-thirds of its business activity is illegal. That’s a big concern for New York State Senator Liz Krueger, known as “Airbnb’s doubter-in-chief.”

Do you have any opinions about this topic? Share it in the comment.

Architecting a Machine Learning System for Risk

Architecting a Machine Learning System for Risk

Online risk mitigation

At Airbnb, we want to build the world’s most trusted community. Guests trust Airbnb to connect them with world-class hosts for unique and memorable travel experiences. Airbnb hosts trust that guests will treat their home with the same care and respect that they would their own. The Airbnb review system helps users find community members who earn this trust through positive interactions with others, and the ecosystem as a whole prospers.

We can mitigate the potential for bad actors to carry out different types of attacks in different ways.

1) Product changes

Many risks can be mitigated through user-facing changes to the product that require additional verification from the user.

2) Anomaly detection

Scripted attacks are often associated with a noticeable increase in some measurable metric over a short period of time.

3) Simple heuristics or a machine learning model based on a number of different variables

Fraudulent actors often exhibit repetitive patterns.

 

Machine Learning Architecture

Different risk vectors can require different architectures. For example, some risk vectors are not time critical, but require computationally intensive techniques to detect. An offline architecture is best suited for this kind of detection. For the purposes of this post, we are focusing on risks requiring realtime or near-realtime action. From a broad perspective, a machine-learning pipeline for these kinds of risk must balance two important goals:

1) The framework must be fast and robust.

That is, we should experience essentially zero downtime and the model scoring framework should provide instant feedback.

2) The framework must be agile.

Since fraud vectors constantly morph, new models and features must be tested and pushed into production quickly.

What do you think about this article? What have you learned from it? If you have any opinions, leave comments below or send us a message.