Code Monkey home page Code Monkey logo

automation's Introduction

Automation

Exploring Human Decision Making

Automation is sometimes seen as replacing humans. However, in most cases only some elements of tasks performed by humans is replaced.

For example, an autopilot does not replace human pilots in an airplane. It merely makes the task of flying easier for human pilots, still requiring vigilant supervision of potential failure modes of automation. For many more years, this state of partial automation will remain in safety critical tasks.

Situation is not very different in deremining and adminstering medication to a patient. Whether it is determining insulin dose, or chemotherapy dose, humans are still involved in the critical decisions, often aided by computers and software.

Therefore, human decision making process under partial automation needs to be examined closely. In addition, understanding the process of human decision making is important even for human behavior not involving automation per se.

Here we will explore this topic using Jupyter notebooks and Python for live interaction with the user. I want to systematically explore surprises that occur along the way. In doing so, I want to point out various method pitfalls relating to human decisions. Sometimes, these pitfalls are present in computer automation methods.

In these examples, I will first describe a problem and the methods used commonly. Then I will describe assumptions underlying the methods and show that it works as intended when used on a simulation that matches the assumptions. Next I will show what happens when used on a more realistic situation simulated using a more complex model. Granted that reality is always more complex than even the complex simulation, it can still show problems encountered in practice.

Topics

  1. Prediction for Automation
  2. Poor Control Inspite of Perfect Prediction
  3. Errors and Uncertainty in Prediction
  4. Interconnted Systems

It seems logical that if you want to control something, you need to measure it and predict it.

For example, if you want to achieve a particular health outcome such as blood glucose level or blood pressure, measuring it, predicting its response to nutrition, exercise and medication seems sufficient. Once you can predict it well, then achieving your goal of controlling it appears a matter of diligent execution by changing the inputs.

We all know from personal experience, that it is easier said than done.

Prediction for automation is different from generic statistical prediction because errors in prediction can have drastic consequences. Moreover, it is common that new modalities of the system (the environment in Reinforcement Learning) are discovered as a direct result of the use of the predictions to change inputs.

automation's People

Contributors

sunilshah avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.