The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

I recently listened to a podcast with the founders of a startup by the name of Aerial (https://aerial.ai) that that is doing real-time location mapping and activity detection using wifi and deep learning.

NOTE: I'm in no way related to this company or the podcast (aside from being an occasional listener).

Episode page: https://twimlai.com/talk/107

Direct episode link: https://feeds.soundcloud.com/stream/393602724-twiml-twiml-ta...

Episode description:

In this episode I’m joined by Michel Allegue and Negar Ghourchian of Aerial.ai. Aerial is doing some really interesting things in the home automation space, by using wifi signal statistics to identify and understand what’s happening in our homes and office environments.

Michel, the CTO, describes some of the capabilities of their platform, including its ability to detect not only people and pets within the home, but surprising characteristics like breathing rates and patterns. He also gives us a look into the data collection process, including the types of data needed, how they obtain it, and how it is parsed. Negar, a senior data scientist with Aerial, describes the types of models used, including semi-supervised, unsupervised and signal processing based models, and how they’ve scaled their platform, and provides us with some real-world use cases.

I'm steadily in the Overcast 1.7x+ camp. The audio processing is nothing short of phenomenal (if a drain on battery on my iPhone 6), and over time, I've adjusted to the point where the speed seems natural and I can spare enough CPU cycles to steer me through my commute.

Other commenters' points that it is also a sort of focus enhancer also ring true to me--I have arrived at the office with nearly 100% retention of what a specific recording was about, especially since I tend to listen to technical podcasts like TWIML[1] (which focuses on stuff I do at work), and I get nothing but value from being able to listen to two podcasts during my one-hour commutes.

Softer stuff like Merlin Mann, Gruber, etc. I will either navigate with the chapter guide or making liberal use of the skip buttons. I seldom (if ever) listen to any kind of podcast at normal speed, except for some language courses.

Even then, I tend to slow down only for languages like Mandarin or Hebrew (which I'm less acquainted with). Spanish, French, etc. (which I know from childhood) tend to be drawn out enough that I can even speed up a little :)

[1]: https://twimlai.com/

Seeing this reminded me of an episode of TWiML&AI [0] (This week in machine learning & AI podcast). It features a very interesting discussion of how Stripe provides an explanation for their black box fraud model to customers when it chooses to block transactions. They basically have created an explanatory model which they train against the black box model and then run against the fraudulent transaction. It outputs the first explanation from a list of human interpretable explanations that would have changed the fraud decision from "Fraud" to "Not Fraud." This is one of the most interesting episodes in the show based on the number of times it has come up in conversation.

In looking up the show notes for that episode, I see that there is also a mention of this paper (in the OP), and that the authors were previously interviewed in TWiML #7 (which I haven't listened to).

0: https://twimlai.com/twiml-talk-73-exploring-black-box-predic...

I happen to agree with this. Have a listen to the last twimlai.com podcast about "deep genomics". From an ML point of view, they're still finding their way, but from an economic point of view two things seem clear. 1) ML + genetics are going to have a massive economic impact. 2) Not a single job will be lost as a result of advancement, as the sector doesn't yet exist. Whenever I hear someone say there won't be any jobs after AI surpasses some threshold, I always ask them how many jobs were lost as a result of railroads? Was it so awful that the Pony Express went out of business after the transcontinental railroad was finished? Or that all those scribes had to quit copying their mistakes over and over again after the printing press came around?