Exam AWS Certified Machine Learning - Specialty topic 1 question 23 discussion

C might be much suitable softmax is to turn numbers into probabilities. https://medium.com/data-science-bootcamp/understand-the-softmax-function-in-minutes-f3a59641e86d

C is right. Softmax function is used for multi-class predictoins

In a multiclass classification problem (such as classifying an image into one of 10 animal categories), the model should output a probability distribution over the classes. The Softmax function achieves this by: Taking the raw scores (logits) from the final dense layer (10 nodes, one per class). Exponentializing each score and normalizing them so they sum to 1, effectively turning them into probabilities.

A. NO - Dropout is to prevent overfitting B. NO - L1 regularization is to prevent overfitting C. YES - Softmax will give probabilities for each class D. NO - Rectified linear units (ReLU) is an activation function

Softmax is the correct answer.

Multiclassification with probabilities is about softmax!

Softmax is for probability distribution

it should be C. Softmax Softmax converts outputs to Probabilites of each classification

absolutely C

Absolute C.

This is as easy a question as you will likely see on the exam, Everyone has the right answer here.

C --> Softmax. Let's go over the alternatives: A. Dropout --> Not really a function, but rather a method to avoid overfitting. It consists of dropping some neurons during the training process, so that the performance of our algorithm does not become very dependent on any single neuron. B. Smooth L1 loss --> It's a loss function, thus a function to be minimized by the entire neural network. It's not an activation function. C. Softmax --> This is the traditional function used for multi-class classification problems (such as classifying an animal into one of 10 categories) D. Rectified linear units (ReLU) --> This activation function is often used on the first and intermediate (hidden) layers, not on the final layer. In any case, it wouldn't make sense to use it for classification because its values can exceed 1 (and probabilities can't)

C, Softmax is the best suitable answer Ref: The softmax function, also known as softargmax[1]:184 or normalized exponential function,[2]:198 is a generalization of the logistic function to multiple dimensions. It is used in multinomial logistic regression and is often used as the last activation function of a neural network to normalize the output of a network to a probability distribution over predicted output classes, based on Luce's choice axiom.

You guys are right, the answer is C since it automatically provides the output with a confidence interval... Relu could be used as well but it needs to be coded in to provide the probabilities https://medium.com/@himanshuxd/activation-functions-sigmoid-relu-leaky-relu-and-softmax-basics-for-neural-networks-and-deep-8d9c70eed91e

Definitely C

Definitely softmax.

Are you sure it is C?` The output should be "[the probability that] the input image belongs to each of the 10 classes." And not the most likely class with the highest probability, which would be the result of softmax layer.