Introduction to Machine Learning Talk

There was an amazing turnout at last night’s DataPhilly meetup (~200 people!). I was completely delighted by the turnout and people’s engagement level. Here are the slides of the talk I gave to set up the evening with a high-level introduction to machine learning.

 

Speaking at DataPhilly February 2016

The next DataPhilly meetup will feature a medley of machine-learning talks, including an Intro to ML from yours truly. Check out the speakers list and be sure to RSVP. Hope to see you there!

Thursday, February 18, 2016

6:00 PM to 9:00 PM

Speakers:

  • Corey Chivers
  • Randy Olson
  • Austin Rochford

Corey Chivers (Penn Medicine)

Abstract: Corey will present a brief introduction to machine learning. In his talk he will demystify what is often seen as a dark art. Corey will describe how we “teach” machines to learn patterns from examples by breaking the process into its easy-to-understand component parts. By using examples from fields as diverse as biology, health-care, astrophysics, and NBA basketball, Corey will show how data (both big and small) is used to teach machines to predict the future so we can make better decisions.

Bio: Corey Chivers is a Senior Data Scientist at Penn Medicine where he is building machine learning systems to improve patient outcomes by providing real-time predictive applications that empower clinicians to identify at risk individuals. When he’s not pouring over data, he’s likely to be found cycling around his adoptive city of Philadelphia or blogging about all things probability and data at bayesianbiologist.com.

Randy Olson (University of Pennsylvania Institute for Biomedical Informatics):

Automating data science through tree-based pipeline optimization

Abstract: Over the past decade, data science and machine learning has grown from a mysterious art form to a staple tool across a variety of fields in business, academia, and government. In this talk, I’m going to introduce the concept of tree-based pipeline optimization for automating one of the most tedious parts of machine learning — pipeline design. All of the work presented in this talk is based on the open source Tree-based Pipeline Optimization Tool (TPOT), which is available on GitHub at https://github.com/rhiever/tpot.

Bio: Randy Olson is an artificial intelligence researcher at the University of Pennsylvania Institute for Biomedical Informatics, where he develops state-of-the-art machine learning algorithms to solve biomedical problems. He regularly writes about his latest adventures in data science at RandalOlson.com/blog, and tweets about the latest data science news at http://twitter.com/randal_olson.

Austin Rochford (Monetate):

Abstract: Bayesian optimization is a technique for finding the extrema of functions which are expensive, difficult, or time-consuming to evaluate. It has many applications to optimizing the hyperparameters of machine learning models, optimizing the inputs to real-world experiments and processes, etc. This talk will introduce the Gaussian process approach to Bayesian optimization, with sample code in Python.

Bio: Austin Rochford is a Data Scientist at Monetate. He is a former mathematician who is interested in Bayesian nonparametrics, multilevel models, probabilistic programming, and efficient Bayesian computation.

A probabilistic justification to carpe diem

There’s a curious thing about unlikely independent events: no matter how rare, they’re most likely to happen right away.

Let’s get hypothetical

You’ve taken a bet that pays off if you guess the exact date of the next occurrence of a rare event (p = 0.0001 on any given day i.i.d). What day do you choose? In other words, what is the most likely day for this rare event to occur?

Setting aside for now why in the world you’ve taken such a silly sounding bet, it would seem as though a reasonable way to think about it would be to ask: what is the expected number of days until the event? That must be the best bet, right?

We can work out the expected number of days quite easily as 1/p = 10000. So using the logic of expectation, we would choose day 10000 as our bet.

Let’s simulate to see how often we would win with this strategy. We’ll simulate the outcomes by flipping a weighted coin until it comes out heads. We’ll do this 100,000 times and record how many flips it took each time.

p0001

The event occurred on day 10,000 exactly 35 times. However, if we look at a histogram of our simulation experiment, we can see that the time it took for the rare event to happen was more often short, than long. In fact, the event occurred 103 times on the very first flip (the most common Time to Event in our set)!

So from the experiment it would seem that the most likely amount of time to pass until the rare event occurs is 0. Maybe our hypothetical event was just not rare enough. Let’s try it again with p=0.0000001, or an event with a 1 in 1million chance of occurring each day.

p0000001

While now our event is extremely unlikely to occur, it’s still most likely to occur right away.

Existential Risk

What does this all have to do with seizing the day? Everything we do in a given day comes with some degree of risk. The Stanford professor Ronald A. Howard conceived of a way of measuring the riskiness of various day-to-day activities, which he termed the micromort. One micromort is a unit of risk equal to p = 0.000001 (1 in a million chance) of death. We are all subject to a baseline level of risk in micromorts, and additional activities may add or subtract from that level (skiing, for instance adds 0.7 micromorts per day).

While minimizing the risks we assume in our day-to-day lives can increase our expected life span, the most likely exact day of our demise is always our next one. So carpe diem!!

Post Script:

Don’t get too freaked out by all of this. It’s just a bit of fun that comes from viewing the problem in a very specific way. That is, as a question of which exact day is most likely. The much more natural way to view it is to ask, what is the relative probability of the unlikely event occurring tomorrow vs any other day but tomorrow. I leave it to the reader to confirm that for events with p < 0.5, the latter is always more likely.

What’s Warren Buffett’s $1 Billion Basketball Bet Worth?

A friend of mine just alerted me to a story on NPR describing a prize on offer from Warren Buffett and Quicken Loans. The prize is a billion dollars (1B USD) for correctly predicting all 63 games in the men’s Division I college basketball tournament this March. The facebook page announcing the contest puts the odds at 1:9,223,372,036,854,775,808, which they note “may vary depending upon the knowledge and skill of entrant”.

Being curious, I thought I’d see what the assumptions were that went into that number. It would make sense to start with the assumption that you don’t know a lick about college basketball and you just guess using a coin flip for every match-up. In this scenario you’re pretty bad, but you are no worse than random. If we take this assumption, we can calculate the odds as 1/(0.5)^63.  To get precision down to a whole integer I pulled out trusty bc for the heavy lifting:

$ echo "scale=50;  1/(0.5^63)" | bc
9223372036854775808.000000

Well, that was easy. So if you were to just guess randomly, your odds of winning the big prize would be those published on the contest page. We can easily calculate the expected value of entering the contest as P(win)*prize, or 9,223,372,036ths of a dollar (that’s 9 nano dollars, if you’re paying attention). You’ve literally already spent that (and then some) in opportunity cost sunk into the time you are spending thinking about this contest and reading this post (but read on, ’cause it’s fun!).

But of course, you’re cleverer than that. You know everything about college basketball – or more likely if you are reading this blog – you have a kickass predictive model that is going to up your game and get your hands into the pocket of the Oracle from Omaha.

What level of predictiveness would you need to make this bet worth while? Let’s have a look at the expected value as a function of our individual game probability of being correct.

buffet1

And if you think that you’re really good, we can look at the 0.75 to 0.85 range:

buffet2

So it’s starting to look enticing, you might even be willing to take off work for a while if you thought you could get your model up to a consistent 85% correct game predictions, giving you an expected return of ~$35,000. A recent paper found that even after observing the first 40 scoring events, the outcome of NBA games is only predictable at 80%. In order to be eligible to win, you’ve obviously got to submit your picks before the playoff games begin, but even at this herculean level of accuracy, the expected value of an entry in the contest plummets down to $785.

Those are the odds for an individual entrant, but what are the chances that Buffet and co will have to pay out? That, of course, depends on the number of entrants. Lets assume that the skill of all entrants is the same, though they all have unique models which make different predictions. In this case we can get the probability of at least one of them hitting it big. It will be the complement of no one winning. We already know the odds for a single entrant with a given level of accuracy, so we can just take the probability that each one doesn’t win, then take 1 minus that value.

buffet3

Just as we saw that the expected value is very sensitive to the predictive accuracy of the participant, so too is the probability that the prize will be awarded at all. If 1 million super talented sporting sages with 80%  game-level accuracy enter the contest, there will only be a slightly greater than 50% chance of anyone actually winning. If we substitute in a more reasonable (but let’s face it, still wildly high) figure for participants’ accuracy of 70%, the chance becomes only 1 in 5739  (0.017%) that the top prize will even be awarded even with a 1 million strong entrant pool.

tl;dr You’re not going to win, but you’re still going to play.

If you want to reproduce the numbers and plots in this post, check out this gist.

Calculating AUC the hard way

The Area Under the Receiver Operator Curve is a commonly used metric of model performance in machine learning and many other binary classification/prediction problems. The idea is to generate a threshold independent measure of how well a model is able to distinguish between two possible outcomes. Threshold independent here just means that for any model which makes continuous predictions about binary outcomes, the conversion of the continuous predictions to binary requires making the choice of an arbitrary threshold above which will be a prediction of 1, below which will be 0.

AUC gets around this threshold problem by integrating across all possible thresholds. Typically, it is calculated by plotting the rate of false positives against false negatives across the range of possible thresholds (this is the Receiver Operator Curve)  and then integrating (calculating the area under the curve). The result is typically something like this:

auc

I’ve implemented this algorithm in an R script (https://gist.github.com/cjbayesian/6921118) which I use quite frequently. Whenever I am tasked with explaining the meaning of the AUC value however, I will usually just say that you want it to be 1 and that 0.5 is no better than random. This usually suffices, but if my interlocutor is of the particularly curious sort they will tend to want more. At which point I will offer the interpretation that the AUC gives you the probability that a randomly selected positive case (1) will be ranked higher in your predictions than a randomly selected negative case (0).

Which got me thinking – if this is true, why bother with all this false positive, false negative, ROC business in the first place? Why not just use Monte Carlo to estimate this probability directly?

So, of course, I did just that and by golly it works.

source("http://polaris.biol.mcgill.ca/AUC.R")
bs<-function(p)
{
 U<-runif(length(p),0,1)
 outcomes<-U<p
 return(outcomes)
}

# Simulate some binary outcomes #
n <- 100
x <- runif(n,-3,3)
p <- 1/(1+exp(-x))
y <- bs(p)

# Using my overly verbose code at https://gist.github.com/cjbayesian/6921118
AUC(d=y,pred=p,res=500,plot=TRUE)

## The hard way (but with fewer lines of code) ##
N <- 10000000
r_pos_p <- sample(p[y==1],N,replace=TRUE)
r_neg_p <- sample(p[y==0],N,replace=TRUE)

# Monte Carlo probability of a randomly drawn 1 having a higher score than
# a randomly drawn 0 (AUC by definition):

rAUC <- mean(r_pos_p > r_neg_p)
print(rAUC)

By randomly sampling positive and negative cases to see how often the positives have larger predicted probability than the negatives, the AUC can be calculated without the ROC or thresholds or anything. Now, before you object that this is necessarily an approximation, I’ll stop you right there – it is.  And it is more computationally expensive too. The real value for me in this method is for my understanding of the meaning of AUC. I hope that it has helped yours too!

Uncertainty matters

In a post I wrote earlier this year, I noted a sentiment expressed in The Economist about understanding and embracing uncertainty.

…recent reforms to the IPCC’s procedures will do little to change its tendency to focus on the areas where there is greater consensus, avoiding the uncertainties which, though unpalatable for scientists, are important to policy. (link)

Which I felt was contrary to the way we, as scientists, speak among ourselves about policy makers. Specifically, that it is they who fear and misunderstand the implications of uncertainty.

This is the same perception which has led to the launch today by the group Sense About Science of a publication titled Making Sense of Uncertainty: Why uncertainty is part of science.

Launching a guide to Making Sense of Uncertainty at the World Conference of Science Journalists today, researchers working in some of the most significant, cutting edge fields say that if policy makers and the public are discouraged by the existence of uncertainty, we miss out on important discussions about the development of new drugs, taking action to mitigate the impact of natural hazards, how to respond to the changing climate and to pandemic threats.

Interrogated with the question ‘But are you certain?’, they say, they have ended up sounding defensive or as though their results are not meaningful. Instead we need to embrace uncertainty, especially when trying to understand more about complex systems, and ask about operational knowledge: ‘What do we need to know to make a decision? And do we know it?’

The report seems to be in line with arguments I have made about uncertainty and decision making as they pertain to ecological research, management, and policy.

Among the contributors to the report is someone who I consider to be among the best when it comes to understanding and communicating uncertainty, David Spiegelhalter. While I haven’t made my way all the way through it yet, it looks like this report will be an informative read for both scientists and policy makers (oh ya, and journalists — can’t forget about them).

Who knows, we might be able to stop the finger pointing and work together in mutual understanding of the importance of uncertainty.

How likely is the NSA PRISM program to catch a terrorist?

Recent revelations about PRISM, the NSA’s massive program of surveillance of civilian communications have caused quite a stir. And rightfully so, as it appears that the agency has been granted warrantless direct access to just about any form of digital communication engaged in by American citizens, and that their access to such data has been growing significantly over the past few years.

Some may argue that there is a necessary trade-off between civil liberties and public safety, and that others should just quit their whining. Lets take a look at this proposition (not the whining part). Specifically, let’s ask: how much benefit, in terms of thwarted would-be attacks, does this level of surveillance confer?

Lets start by recognizing that terrorism is extremely rare. So the probability that an individual under surveillance (and now everyone is under surveillance) is also a terrorist is also extremely low. Lets also assume that the neck-beards at the NSA are fairly clever, if exceptionally creepy. We assume that they have devised an algorithm that can detect ‘terrorist communications’ (as opposed to, for instance, pizza orders) with 99% accuracy.

P(+ |  bad guy) = 0.99

A job well done, and Murica lives to fight another day. Well, not quite. What we really want to know is: what is the probability that they’ve found a bad guy, given that they’ve gotten a hit on their screen? Or,

P(bad guy | +) =??

Which is quite a different question altogether. To figure this out, we need a bit more information. Recall that bad guys (specifically terrorists) are extremely rare, say on the order of one in a million (this is a wild over estimate with the true rate being much lower, of course – but lets not let that stop us). So,

P(bad guy) = 1/1,000,000

Further, lets say that the spooks have a pretty good algorithm that only comes up falsely positive (ie when the person under surveillance is a good guy) one in one hundred times.

P(+ |  good guy) = 0.01

And now we have all that we need. Apply a little special Bayes sauce:

P(bad guy | +) = P(+ | bad guy) P(bad guy)  /  [ P(+ |  bad guy) P(bad guy) + P(+ |  good guy) P(good guy) ]

and we get:

P(bad guy | +) = 1/10,102

That is, for every positive (the NSA calls these ‘reports’) there is only a 1 in 10,102 chance (using our rough assumptions) that they’ve found a real bad guy.

UPDATE: While former NSA analyst turned whistle blower William Binney thinks this is a plausible estimate, the point here is not that this is the ‘correct probability‘ involved (remember that we based our calculations on very rough assumptions). The take away message is simply that whenever the rate of an event of interest is extremely low, even a very accurate test will fail very often.

UPDATE 2: The Wall Street Journal’s Numbers Guy has written a piece on this in which several statisticians and security experts respond.

UPDATE 3: If you can read German, a reader reached me to point out that Der Spiegel technology section picked up the story.

Big brother is always watching, but he’s still got a needle in a haystack problem.

Big Brother 11

The television series doesn’t have this problem. On the show, they’re all bad guys.