Problem 58917. Neural Net: Back Propagation of Quad Output with Bias, ReLU

• Created by Richard Zapor

×

Like (0)

Difficulty:

(76)

Rate

• Solve Later   
• Add To GroupÂ

This challenge is to return the WH_delta and WP_delta, given X, WH, WP, EPY using ReLU on the hidden layer and Softmax on the output layer. Test Cases will accumulate dWP and dWH to solve neural nets for Counter, Subtractor,Mux. Test Cases will have four output cases. ReLU performs well on multiple output cases.

[dWP,dWH]=Neural_Back_Propagation_ReLU(X,WH,WP,EPY)

The matlab Latex code for making the Back Propagation chart included in template.

Solve

Solution Stats

100.0% Correct | 0.0% Incorrect

• 6 Solutions
• 5 Solvers

Last Solution submitted on Feb 08, 2024

Last 200 Solutions

Solution Comments

ANNOUNCEMENT ×

Registration Now Open for MathWorks AUTOMOTIVE CONFERENCE 2025

Hello Community, We're excited to announce that registration is now open for the...

Problem Recent Solvers5

Suggested Problems

• Create a function handle that reverses the input arguments of another function handle

  152 Solvers

• Return the first and last characters of a character array

  10637 Solvers

• Construct an index vector from two input vectors in vectorized fashion

  404 Solvers

• Make a 1 hot vector

  517 Solvers

• Possible Outcomes of American Roulette

  119 Solvers

More from this Author308

• ICFP 2022 - Optimal RGB for a region

  3 Solvers

• ICFP2024 011:Lambdaman Small programmed solutions

  3 Solvers

• Create formatted value string Cell array of a vector

  27 Solvers

• Primes for Large N: 2^30, System Memory Limit

  25 Solvers

• R2012b atan in Degrees

  71 Solvers