Skip to main content

When the ATM fails to read my checks

Happy Lunar New Year! And greetings to Orlando people who are coming to my dataviz seminar this morning. *** What’s going on with digit recognition, one of the signature applications of machine learning? Before self-driving cars, before image recognition, before machine translation, there was digit recognition: computers are trained to read and recognize hand-written numbers. This problem shares several of the key components of problems tailor-made for machine learning methods: The correct answer is unambiguous for each item (i.e. image of a digit). The author of the digit has a particular number in mind. The range of possible answers across all items is finite. In a decimal system, each image can only be one of 0, 1, 2, ... , 9. The end-user only cares about how accurately the digit can be predicted. Causality is not of interest here. A massive dataset of labeled images, i.e. images that have been correctly recognized, used to train computers is easily obtained. Live application generates more data, which feeds back into the system in a positively-reinforcing manner. This digit-recognition technology is widely deployed. Check deposit at the ATM machine is one obvious example. In 2016, about 16 billion checks are deposited in the U.S. (source). So what’s wrong with the current state of art? This snapshot I took at an ATM illustrates the problem: Recently, I noticed that the ATM has refused to recognize the digits on several checks, asking me to enter the amounts manually instead. From this evidence, I infer the following: Still after these years, the error rate is higher than these banks could absorb. Assuming 10 billion checks read each year at ATMs, even a 0.01% error rate amounts to 1 million errors per year, or about 2,800 errors per day! Banks would rather err on the side of caution – when in doubt, ask users to enter the amount. This behavior implies humans make fewer errors than machines, even after including mischief as a source of human error. What would a teller do if s/he can't make out the scripted digits? The human would look at the handwritten words "six thousand," solving the problem. Apparently, the ATM does not have handwriting recognition technology, or perhaps its accuracy rate is not high enough. It's a harder problem, though of a similar nature. *** Why are the banks risk averse? As a victim of one of these errors, I think I understand. Last year, I spent four or six weeks chasing after $20. In this case, the machine read the 2 as a 0. I didn't catch the mistake while at the ATM, but later noticed it on the bank statement. I learned that convenience comes at a price. The bank's process to verify the amount and correct the mistake is convoluted. It's like missing that exit on the highway, and you now have to go five miles before the next exit. It's a pain for the bank as well as for the client.  

from Big Data, Plainly Spoken (aka Numbers Rule Your World) http://bit.ly/2Tz02iP
via IFTTT

Comments

Popular posts from this blog

Zooming in on maps with sf and ggplot2

(This article was first published on r-bloggers – WZB Data Science Blog , and kindly contributed to R-bloggers) When working with geo-spatial data in R, I usually use the sf package for manipulating spatial data as Simple Features objects and ggplot2 with geom_sf for visualizing these data. One thing that comes up regularly is “zooming in” on a certain region of interest, i.e. displaying a certain map detail. There are several ways to do so. Three common options are: selecting only certain areas of interest from the spatial dataset (e.g. only certain countries / continent(s) / etc.) cropping the geometries in the spatial dataset using sf_crop() restricting the display window via coord_sf() I will show the advantages and disadvantages of these options and especially focus on how to zoom in on a certain point of interest at a specific “zoom level”. We will see how to calculate the coordinates of the display window or “bounding box” around this zoom point. Loading t...

Controlling legend appearance in ggplot2 with override.aes

[This article was first published on Very statisticious on Very statisticious , and kindly contributed to R-bloggers ]. (You can report issue about the content on this page here ) Want to share your content on R-bloggers? click here if you have a blog, or here if you don't. In ggplot2 , aesthetics and their scale_*() functions change both the plot appearance and the plot legend appearance simultaneously. The override.aes argument in guide_legend() allows the user to change only the legend appearance without affecting the rest of the plot. This is useful for making the legend more readable or for creating certain types of combined legends. In this post I’ll first introduce override.aes with a basic example and then go through three additional plotting scenarios to how other instances where override.aes comes in handy. Table of Contents R packages Introducing override.aes Adding a guides() layer Using the guide argument in scale_*() Changing multiple aesthetic par...