• Posts
  • RSS
  • ◂◂RSS
  • Contact

  • Exercises

    July 14th, 2012
    learning  [html]
    The naive way of learning to do something is just to attempt it repeatedly. This works remarkably well for people given how hard it is to teach computers to do the same tasks, but it's not how we usually teach people to do things. Instead we've found you learn faster and at lower cost if you do exercises.

    With the naive approach, if we wanted children to be good at handling checkout at stores we would just have them do it. They could pretty quickly learn, under supervision, where to put money that someone gave them, but giving correct change would be hard. The supervisor could simply stand there, pointing out every time they gave incorrect change, but that would lead to slow learning and a lot of supervisor time. It might be better to use exercises: spend some time subtracting typical prices from typical combinations of offered bills. Even this might be too much for ideal practice: what change to give for a $20 when someone gives you $3.52 is kind of hard. So the supervisor recognizes that the semi-abstract skill of giving change has a more abstract subskill of simple subtraction and has the child practice that. You can go another step: 20.00 - 3.52 may even be too hard, so we do exercises in subtracting low whole numbers 10-3, 5-2, 6-1, and build from there.

    The risk of an exercise, however, is that it may not match very well to the final goal. I remember two strategies in physics that were very helpful in solving problems:

    1. What have we been recently learning in class? How might that apply here?
    2. Have I used all the numbers they gave me?
    These turn out to be great for a lot of physics problems, but they stop helping once you try to use physics to answer real questions.

    In the same class, I started using a strategy of assigning variables to all the given numbers, and solving to problem with variables, substituting numbers for the variables only at the end once I'd done all of the algebraic manipulation. I found this helped some in reducing errors and was less writing, so I liked it, but no one else was interested and the teacher thought it was strange. Which is unfortunate: this approach is good practice for automation, a big part of the final goal for anyone actually using physics.

    Designing good exercises means understanding what will be useful for larger scale exercises and skills. This is hard, and is why there are so often exercises that don't really teach the skills they're intended to.

    Comment via: google plus, facebook

    Recent posts on blogs I like:

    What should we do about network-effect monopolies?

    Many large companies today are software monopolies that give their product away for free to get monopoly status, then do horrible things. Can we do anything about this?

    via benkuhn.net July 5, 2020

    More on the Deutschlandtakt

    The Deutschlandtakt plans are out now. They cover investment through 2040, but even beforehand, there’s a plan for something like a national integrated timetable by 2030, with trains connecting the major cities every 30 minutes rather than hourly. But the…

    via Pedestrian Observations July 1, 2020

    How do cars fare in crash tests they're not specifically optimized for?

    Any time you have a benchmark that gets taken seriously, some people will start gaming the benchmark. Some famous examples in computing are the CPU benchmark specfp and video game benchmarks. With specfp, Sun managed to increase its score on 179.art (a su…

    via Posts on Dan Luu June 30, 2020

    more     (via openring)


  • Posts
  • RSS
  • ◂◂RSS
  • Contact