By using this site, you agree to the Privacy Policy and Terms of Use.
Accept
World of SoftwareWorld of SoftwareWorld of Software
  • News
  • Software
  • Mobile
  • Computing
  • Gaming
  • Videos
  • More
    • Gadget
    • Web Stories
    • Trending
    • Press Release
Search
  • Privacy
  • Terms
  • Advertise
  • Contact
Copyright © All Rights Reserved. World of Software.
Reading: What Happens if You Remove ReLU From a Deep Neural Network? | HackerNoon
Share
Sign In
Notification Show More
Font ResizerAa
World of SoftwareWorld of Software
Font ResizerAa
  • Software
  • Mobile
  • Computing
  • Gadget
  • Gaming
  • Videos
Search
  • News
  • Software
  • Mobile
  • Computing
  • Gaming
  • Videos
  • More
    • Gadget
    • Web Stories
    • Trending
    • Press Release
Have an existing account? Sign In
Follow US
  • Privacy
  • Terms
  • Advertise
  • Contact
Copyright © All Rights Reserved. World of Software.
World of Software > Computing > What Happens if You Remove ReLU From a Deep Neural Network? | HackerNoon
Computing

What Happens if You Remove ReLU From a Deep Neural Network? | HackerNoon

News Room
Last updated: 2026/03/05 at 4:30 PM
News Room Published 5 March 2026
Share
What Happens if You Remove ReLU From a Deep Neural Network? | HackerNoon
SHARE

Removed ReLU from a 5-layer PyTorch MLP. The model trained without errors, loss decreased every epoch, and it still hit 91.8% on MNIST — matching single-layer logistic regression exactly. Four hidden layers with 575K parameters added zero expressive power.
The gradient data was the unexpected part. Early layers didn’t vanish — they just received proportionally less signal. By epoch 10, fc5 had a gradient norm of 1.37 while fc1 sat at 0.37. The optimizer was concentrating updates at the output end because that’s where the matrix product chain is shortest. The network quietly specialized its final layer to do all the classification work and treated the rest as a passive linear projection.
Depth made it worse, not neutral. A 10-layer linear network scored 90.93% vs 92.00% for a 1-layer one — same expressive ceiling, harder optimization landscape.
Every non-linear activation tested (Sigmoid, Tanh, ReLU, Leaky ReLU, GELU) landed between 97.4% and 98.2%. The gap between any activation and none was ~6 points. The choice of which activation barely mattered. The choice of whether to use one was everything.
The real job of ReLU isn’t gradient management. It’s making depth mean something at all.

Full experiment: 5 scripts, 3 seeds per config, 21,523 seconds on CPU. Code at : https://github.com/Emmimal/relu-experiment

Sign Up For Daily Newsletter

Be keep up! Get the latest breaking news delivered straight to your inbox.
By signing up, you agree to our Terms of Use and acknowledge the data practices in our Privacy Policy. You may unsubscribe at any time.
Share This Article
Facebook Twitter Email Print
Share
What do you think?
Love0
Sad0
Happy0
Sleepy0
Angry0
Dead0
Wink0
Previous Article Amazon.com is up and down, with login errors and prices not loading Amazon.com is up and down, with login errors and prices not loading
Next Article Diablo 4 Warlock Gameplay: Check Out the 4 Subclasses of the Apocalypse Diablo 4 Warlock Gameplay: Check Out the 4 Subclasses of the Apocalypse
Leave a comment

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Stay Connected

248.1k Like
69.1k Follow
134k Pin
54.3k Follow

Latest News

Solving the Payroll Onboarding Headache: How NE2NE Helps Teams Move Faster | HackerNoon
Solving the Payroll Onboarding Headache: How NE2NE Helps Teams Move Faster | HackerNoon
Computing
Today's NYT Strands Hints, Answer and Help for March 6 #733 – CNET
Today's NYT Strands Hints, Answer and Help for March 6 #733 – CNET
News
Testing Strategies for LLM-Generated Web Development Code | HackerNoon
Testing Strategies for LLM-Generated Web Development Code | HackerNoon
Computing
By any other name: A18 Pro is just as good a Mac chip as M4 for most
By any other name: A18 Pro is just as good a Mac chip as M4 for most
News

You Might also Like

Solving the Payroll Onboarding Headache: How NE2NE Helps Teams Move Faster | HackerNoon
Computing

Solving the Payroll Onboarding Headache: How NE2NE Helps Teams Move Faster | HackerNoon

7 Min Read
Testing Strategies for LLM-Generated Web Development Code | HackerNoon
Computing

Testing Strategies for LLM-Generated Web Development Code | HackerNoon

1 Min Read
System76 Comments On Recent Age Verification Laws
Computing

System76 Comments On Recent Age Verification Laws

3 Min Read
Production Observability for Multi-Agent AI (with KAOS + OTel + SigNoz) | HackerNoon
Computing

Production Observability for Multi-Agent AI (with KAOS + OTel + SigNoz) | HackerNoon

28 Min Read
//

World of Software is your one-stop website for the latest tech news and updates, follow us now to get the news that matters to you.

Quick Link

  • Privacy Policy
  • Terms of use
  • Advertise
  • Contact

Topics

  • Computing
  • Software
  • Press Release
  • Trending

Sign Up for Our Newsletter

Subscribe to our newsletter to get our newest articles instantly!

World of SoftwareWorld of Software
Follow US
Copyright © All Rights Reserved. World of Software.
Welcome Back!

Sign in to your account

Lost your password?