Linear Digressions is a podcast about Machine Learning and Data Science
Mountain View, CA
Joined October 2015
- Tweets337
- Following10
- Followers4.1K
- Likes179
After over 5 amazing years of podcasting, we're hanging up our microphones--it's with bittersweetness that we say that this week's episode will be the last of Linear Digressions. It's been so much fun joining you each week, and we will miss you all. Thanks for listening! 🤓❤️👋
Data scientist and PhD student in criminology @ZachDrakeTweets is on the pod this week, unpacking statistical pitfalls, challenges detangling cause and effect, and data-driven criminal justice policy from an expert perspective
This week: explaining an algorithm that's used to sentence defendants. It's harsher on black defendants than whites. But the algorithm isn't buggy, it's the patterns it learns from replicating racist policing patterns that are broken. It's the system. #BlackLivesMatter
[Katie here] We're doing a rerun this week. I didn't plan to, but events of last week and this weekend make podcasting seem less important than usual. For folks who've asked over the years how they can support the podcast: donate to a bail fund or food pantry. #BlackLivesMatter
This week's episode features AI expert Prof. Stuart Russell, in a discussion about the risky path that AI is currently on and what changes we should make now for our long-term safety and happiness. AI is powerful stuff, folks--let's not mess this up.
This week's episode: the ups and downs, the ins and outs (mostly ins) of working from home. What are you all doing to stay sane and productive if you're suddenly working from home full time right now?
One thing that’s extremely important to understand, in order to fight covid-19 as effectively as possible, is how the virus spreads and especially how much of the spread of the disease comes from carriers who are experiencing no or mild symptoms but are contagious anyway 1/3
This episode digs into the epidemiological model that was published in Science this week—this model finds that the data suggests that the majority of carriers of the coronavirus, 80-90%, do not have a detected disease
lineardigressions.com/episod…
2/3
This has big implications for the importance of social distancing of a way to get the pandemic under control and explains why a more comprehensive testing program is critical for the United States 3/3
Here's the paper: science.sciencemag.org/conte…
Now seems like a good time to re-release an episode where, among other things, we cover this paper: ncbi.nlm.nih.gov/pmc/article…
What does this mean? STAY HOME RIGHT NOW, as much as you can. Practice good hygiene. It doesn't matter if you think you'll be fine (although even young and healthy people are ending up in ICUs), the network effects of your actions are huge for others.
This week we dig into the Poisson distribution, which can be used to model common everyday processes like army deaths from horse kicks and supernovas
lineardigressions.com/episod…
In 1997, Clayton Christensen wrote the Innovator's Dilemma as an attempt to answer a simple question: why do industry-leading firms so consistently miss the boat on disruptive technologies? 1/
If we think of data science initiatives as a special flavor of innovation initiatives (which they often are!), we (data scientists) can tap into a rich vein of business literature on how to keep the little innovation engine from being swept away by business as usual. 3/
So join us this week as we unpack The Innovator's Dilemma, and think about what it foretells for data scientists working in legacy market-leading firms. 4/4
lineardigressions.com/episod…
Traditional A/B tests assume that whether or not one person got a treatment has no effect on the experiment outcome for another person. But that’s not a safe assumption, especially when there are network effects (like in almost any social context, for instance!)...
From news feeds in LinkedIn to disentangling herd immunity from individual immunity in vaccine studies, indirect (i.e. network) effects in experiments can be just as big as, or even bigger than, direct (i.e. individual effects)...
And this (SUTVA and how to deal with it) is what we talk about this week on the podcast.
Relevant links:
hanj.cs.illinois.edu/pdf/www…
ncbi.nlm.nih.gov/pmc/article…
lineardigressions.com/episod…