AI News, BOOK REVIEW: Why not just dump the neural networks and deep learning? [closed]
- On Monday, June 4, 2018
- By Read More
Why not just dump the neural networks and deep learning? [closed]
As the comments to your question point out, there are a lot of people working on finding something better.
I would though like to answer this question by expanding the comment left by @josh All models are wrong but some are useful (Wiki) The above statement is a general truth used to describe the nature of statistical models.
Using a number of observations, we can fit a model to give us an approximate value for a dependent variable given any value(s) for the independent variable(s).
Deviations from our model (as can be seen in the image above) appear random, some observations are below the line and some are above, but our regression line shows a general correlation.
For example, imagine watching cars as they drove through a junction where they must turn either left or right to continue, the cars turn in no particular pattern.
In reality they are probably heading somewhere specific for a specific reason, and without attempting to stop each car to ask them about their reasoning, we can only describe their actions as random.
Where we are able to fit a model with minimal deviation, how certain can we be that an unknown, unnoticed or immeasurable variable wont at some point throw our model?
As you mention, there's an incomprehensible number of solutions which could fit the data using ANNs, but is this number really any different to the amount of solutions we would need to develop ourselves through trial and error?
- On Tuesday, June 18, 2019
How to Predict Stock Prices Easily - Intro to Deep Learning #7
We're going to predict the closing price of the S&P 500 using a special type of recurrent neural network called an LSTM network. I'll explain why we use ...
SAS Enterprise Miner: Impute, Transform, Regression & Neural Models
Chip Robie of SAS presents the fourth in a series of six "Getting Started with SAS Enterprise Miner ..
TEDxZurich - Molly Crockett - Drugs and morals
Does your sense of fairness depend on what you ate for breakfast? Can Prozac influence your judgment of what is right or wrong? How can we encourage ...
1. What is Computation?
MIT 6.0001 Introduction to Computer Science and Programming in Python, Fall 2016 View the complete course: Instructor: Dr. Ana ..
Oprah Winfrey Presents: O's Little Book of Happiness (The Editor's Best Collection - Full Audiobook)
Watch Next ▻ "PRESENCE, POWER & PROFIT" ▻ Oprah Winfrey Presents: O's Little Book of Happiness ..
Yelawolf - American You
Pre-order the album Love Story now On iTunes: Google Play: Amazon MP3: .
The Truth About Single Moms
Single Mothers are talked about in glowing terms for their hard-work, determination and fortitude - but what is the truth about single moms? Motherhood has ...
Azure Machine Learning: Predict Who Survives the Titanic - Jennifer Marsman - Duo Tech Talk
Interested in doing machine learning in the cloud? In this demo-heavy talk, I will set the stage with some information on the different types of machine learning ...
Yelawolf - Daddy's Lambo
Sign up for updates: Music video by Yelawolf performing Daddy's Lambo. (C) 2011 DGC Records Best of Yelawolf: ..