AI News, The Seductive Trap of Black-Box Machine Learning

The Seductive Trap of Black-Box Machine Learning

For as long as I have been participating in data mining and machine learning competitions, I have thought about automating my participation.

When working on a dataset, I typically spend a disproportionate amount of time thinking about algorithm tuning and running tuning experiments.

I am prone to performing post-competition analysis and regret my allocation of time, almost always thinking that more time could be put into feature engineering, exploring different models and hypotheses, and blending.

system where model running, tuning and result blending is automated and human effort is focused on defining different perspectives on the problem.

The gist of the pipeline design looks like the following: The system is focused on classification and regression problems, because most problems I look at can be reduced to a problem in this form.

Automatic views can be created such as features from feature selection algorithms and transforms such as standardization, normalization, squaring, etc.

As stated, the pipeline design is intended to let the computer(s) focus on automating the procedures allowing the operator to focus on coming up with new perspectives on the problem at hand.

Grid-searching and random searching of algorithm parameters can add new configurations en-mass or as the results of the searching process.

My vision is to have the framework as public open source and quickly accrue algorithms and reports at all levels exploiting multiple machine learning libraries and distributing computing infrastructure where required.

I think the business friendly versions of this vision already exists, it is commoditized machine learning or machine learning as a service (MLaaS).

At best the results of a competition can inject new ideas for methods and processes into the business, but the actual predictions and even the actual models are throw-away.

Automatic programming can give you a program that solves a well defined problem, but you have no idea how ugly that program is under the covers, and you probably don’t want to know.

The very idea of this is repulsive to programmers, for the very same idea that a magic black box machine learning system is repulsive to a machine learning practitioner (data scientist?).

Cognition: How Your Mind Can Amaze and Betray You - Crash Course Psychology #15

You can directly support Crash Course at Subscribe for as little as $0 to keep up with everything we're doing. Also, if you ..

Google's DeepMind AI Just Taught Itself To Walk

Google's artificial intelligence company, DeepMind, has developed an AI that has managed to learn how to walk, run, jump, and climb without any prior guidance ...

How It Works: Design Thinking

Trying to solve a problem or find better ways of getting work done? Get familiar with IBM Design Thinking and Agile. For more information on IBM Design ...

Game Theory: The Science of Decision-Making

With up to ten years in prison at stake, will Wanda rat Fred out? Game theory is looking at human interactions through the lens of mathematics. Hosted by: Hank ...

How to Never Run out of Things to Say - Keep a Conversation Flowing!

Use my FREE 27 Confidence-Boosting Hacks: Want my TOP 10 book list?: Get a girl to like .

6. Monte Carlo Simulation

MIT 6.0002 Introduction to Computational Thinking and Data Science, Fall 2016 View the complete course: Instructor: John Guttag ..

The RSA Encryption Algorithm (1 of 2: Computing an Example)

epic conway's game of life

Super important. How we will win the fight against aging using science: Here the program I used in the video to run the Game of ..

Computational Thinking: Algorithm Design

This video introduces the concept of Algorithm Design in Computational Thinking. It is part of a short course to introduce Middle and High School teachers to ...

But what *is* a Neural Network? | Chapter 1, deep learning

Subscribe to stay notified about new videos: Support more videos like this on Patreon: Special .