wasshuber

wasshuber

Programming Machine Learning: Help: weird results I don't understand

I encountered something that I can’t explain. Any help, tips, or explanations would be great.

I followed the one hidden layer example with 100 nodes and sigmoid activation function. Works great and I can get to 98.6% accuracy with a learning rate of 1.0, a batch size of 1000, and 100 epochs.

I then decided to exchange the sigmoid activation function with the ReLU. This is not done in the book at this point but it is easy enough to program the ReLU and its derivative. Here is the Python code I used:

def relu(z):
    return np.maximum(0.0,z)
def relu_gradient(z):
    return (z > 0)*1

Works fine as long as one reduces the learning rate which I did reduce to 0.1. It reaches about the same level of accuracy as with the sigmoid. I then did one insignificant change in the gradient of the ReLU. Instead of z > 0 I wrote z >= 0. So the code for the gradient was now:

def relu_gradient(z):
    return (z >= 0)*1

This I thought should not make any difference because how often would z be exactly zero? How often would the weighted sum of all inputs in the floating point format be exactly zero? Perhaps never. Even if it is zero occasionally it should hardly make any big difference. But to my surprise, it makes a profound difference. I can only get to about 95%. Why? Why is there almost 4% difference in accuracy for this insignificant change? There must be something weird happening.

I tried this several times to rule out that somehow the random initialization was unusual. I tried it with different learning rates and different batch sizes. None made any difference in the result. I checked for dead neurons. Found none. If somebody can tell me what is going on here I would really appreciate it.

Most Liked

wasshuber

wasshuber

Turns out it was a bug. Using the nomenclature of the book I was feeding h into the gradient function when I should have fed a into it. With the >= comparison this made all the gradients 1 and thus it acted like the linear activation function. (The linear activation function does produce about 94% accuracy.) Properly using the gradient function produces the expected results. It doesn’t matter if one uses > or >=.

I am happy I found this bug. But this is also part of why your book is so great. Programming it yourself forces one to understand the little details and allows one to change and modify the algorithms at the very core, which leads to much deeper understanding of how this all works.

Here is an insight that my experimentation produced. I tested a bunch of different activation functions including weird piecewise linear ones, periodic ones with sin and cos, combinations thereof etc. It surprised me that many work just as good as ReLU or sigmoid with a single hidden layer. (I intend to extend this experimentation to multiple hidden layers.) For example, it is kind of shocking at first that the absolute-value-function works just as good as ReLU. This kind of makes sense in the biological case. A neuron being a cell would not be completely identical to its neighbor neuron. Neurons in nature would certainly have different activation functions. Perhaps not as different as I experimented with but they would perhaps be noisy and distorted versions of sigmoid or ReLU. It doesn’t matter, it still works fine.

Further, this makes me wonder if perhaps that variation in activation functions in nature is a benefit. I am wondering if folks have tried to make nets where each activation function of each neuron is different. Perhaps that confers a training advantage to the network because not everything behaves in exactly the same way? I will try to explore this question. But first I need to extend the code to allow for multiple hidden layers.

This is one critique I have to make. In my opinion, it would have been better to go further with the code and extend it to multiple hidden layers than to switch to libraries. The point of the book is programming it yourself to allow full unmitigated experimentation. I would have added one or two chapters to extend the code further even if that would have meant leaving out libraries altogether. Numpy should be fast enough to explore multilayer networks on a single average computer.

Where Next?

Popular Pragmatic Bookshelf topics Top

ianwillie
Hello Brian, I have some problems with running the code in your book. I like the style of the book very much and I have learnt a lot as...
New
edruder
I thought that there might be interest in using the book with Rails 6.1 and Ruby 2.7.2. I’ll note what I needed to do differently here. ...
New
fynn
This is as much a suggestion as a question, as a note for others. Locally the SGP30 wasn’t available, so I ordered a SGP40. On page 53, ...
New
jskubick
I’m under the impression that when the reader gets to page 136 (“View Data with the Database Inspector”), the code SHOULD be able to buil...
New
brunogirin
When installing Cards as an editable package, I get the following error: ERROR: File “setup.py” not found. Directory cannot be installe...
New
brunogirin
When running tox for the first time, I got the following error: ERROR: InterpreterNotFound: python3.10 I realised that I was running ...
New
dsmith42
Hey there, I’m enjoying this book and have learned a few things alredayd. However, in Chapter 4 I believe we are meant to see the “>...
New
jonmac
The allprojects block listed on page 245 produces the following error when syncing gradle: “org.gradle.api.GradleScriptException: A prob...
New
kolossal
Hi, I need some help, I’m new to rust and was learning through your book. but I got stuck at the last stage of distribution. Whenever I t...
New
mert
AWDWR 7, page 152, page 153: Hello everyone, I’m a little bit lost on the hotwire part. I didn’t fully understand it. On page 152 @rub...
New

Other popular topics Top

AstonJ
What chair do you have while working… and why? Is there a ‘best’ type of chair or working position for developers?
New
brentjanderson
Bought the Moonlander mechanical keyboard. Cherry Brown MX switches. Arms and wrists have been hurting enough that it’s time I did someth...
New
AstonJ
Curious to know which languages and frameworks you’re all thinking about learning next :upside_down_face: Perhaps if there’s enough peop...
New
Margaret
Hello everyone! This thread is to tell you about what authors from The Pragmatic Bookshelf are writing on Medium.
1147 28379 760
New
PragmaticBookshelf
Author Spotlight Rebecca Skinner @RebeccaSkinner Welcome to our latest author spotlight, where we sit down with Rebecca Skinner, auth...
New
hilfordjames
There appears to have been an update that has changed the terminology for what has previously been known as the Taskbar Overflow - this h...
New
PragmaticBookshelf
Develop, deploy, and debug BEAM applications using BEAMOps: a new paradigm that focuses on scalability, fault tolerance, and owning each ...
New
AstonJ
If you’re getting errors like this: psql: error: connection to server on socket “/tmp/.s.PGSQL.5432” failed: No such file or directory ...
New
AstonJ
This is cool! DEEPSEEK-V3 ON M4 MAC: BLAZING FAST INFERENCE ON APPLE SILICON We just witnessed something incredible: the largest open-s...
New
Fl4m3Ph03n1x
Background Lately I am in a quest to find a good quality TTS ai generation tool to run locally in order to create audio for some videos I...
New

Sub Categories: