#0080

L2 Regularized Loss

MediumML18 min20 XP

Problem

Given base_loss, weights, and lambda_value, return base_loss + lambda_value * sum(weight squared). Do not penalize bias separately; weights contains only penalized weights.

Why This Matters

Regularization discourages overly large weights and can improve generalization when models overfit.

Function Signature

def l2_loss(base_loss, weights, lambda_value):

Examples

Example 1

Inputbase_loss=2, weights=[3,4], lambda=0.1

Output4.5

Penalty is 9 + 16 = 25, so add 2.5.

Constraints

weights can be empty.
lambda_value can be zero.
Return a number.

CodePython

def l2_loss(base_loss, weights, lambda_value):
    """
    Input example: base_loss=2, weights=[3,4], lambda=0.1
    Return: 4.5
    Return the value from the function so the tests can compare it. Do not print unless the prompt asks.
    """
    # write your code here
    return base_loss

def l2_loss(base_loss, weights, lambda_value):
    """
    Input example: base_loss=2, weights=[3,4], lambda=0.1
    Return: 4.5
    Return the value from the function so the tests can compare it. Do not print unless the prompt asks.
    """
    # write your code here
    return base_loss

Visible browser tests run here when available.

Testcases1 visible / 3 hidden categories

Two weights

Input2, [3,4], 0.1

Expected4.5

2 + 0.1 * 25.

Hidden Test Categories

zero lambdaempty weightsnegative weights