- A
Sigmoid
Why wrong: Sigmoid also saturates and causes vanishing gradients.
- B
Softmax
Why wrong: Softmax is used only in output layers for multi-class probabilities.
- C
ReLU
ReLU is non-saturating and helps mitigate vanishing gradients.
- D
Linear
Why wrong: Linear activation reduces network to a linear model, losing expressiveness.
Question 85 of 1,000
Activation Functions: ReLU vs Tanh for Vanishing Gradients
This AI0-001 practice question tests your understanding of machine learning and deep learning. Read the scenario carefully and evaluate each option against the stated constraints before committing to an answer. After answering, compare your reasoning against the explanation and wrong-answer breakdown below. Once you have made your selection, read the full explanation to reinforce the concept and understand why each distractor is designed to mislead on exam day.
A deep learning model for sentiment analysis uses a softmax output layer. The hidden layers currently use tanh activation. Which activation function should replace tanh to mitigate vanishing gradients in deeper networks?
Quick Answer
The answer is ReLU. Unlike tanh, which saturates at extreme values and produces near-zero gradients that stall learning in deeper layers, ReLU’s output is unbounded for positive inputs, allowing gradients to flow freely and directly addressing the vanishing gradient problem. This concept is central to the CompTIA AI+ AI0-001 exam, where you must recognize that activation functions to mitigate vanishing gradients must avoid saturation; a common trap is assuming tanh’s zero-centered output is always superior, but in deep networks its saturating tails still cause gradient decay. A reliable memory tip is “ReLU lets the gradient through—tanh slams the door.”
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
ReLU
ReLU (Rectified Linear Unit) is correct because it outputs zero for negative inputs and a positive linear slope for positive inputs, which avoids the saturation problem of tanh. In deeper networks, tanh gradients can vanish as activations approach ±1, slowing or halting learning. ReLU's non-saturating nature keeps gradients flowing for positive inputs, mitigating the vanishing gradient problem.
Key principle: Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Sigmoid
Why it's wrong here
Sigmoid also saturates and causes vanishing gradients.
- ✗
Softmax
Why it's wrong here
Softmax is used only in output layers for multi-class probabilities.
- ✓
ReLU
Why this is correct
ReLU is non-saturating and helps mitigate vanishing gradients.
Related concept
Read the scenario before looking for a memorised answer.
- ✗
Linear
Why it's wrong here
Linear activation reduces network to a linear model, losing expressiveness.
Common exam traps
Common exam trap: answer the scenario, not the keyword
Candidates often mistakenly believe that any non-linear activation works equally well in deep networks, but the trap is that they may choose sigmoid because it is non-linear, ignoring its saturation-induced vanishing gradient problem in deeper architectures.
Trap categories for this question
Command / output trap
Softmax is used only in output layers for multi-class probabilities.
Detailed technical explanation
How to think about this question
ReLU introduces sparsity by zeroing out negative inputs, which can lead to 'dead neurons' if a large gradient forces weights into a region where the neuron never activates. In practice, variants like Leaky ReLU or Parametric ReLU are used to address this. For sentiment analysis, ReLU's computational efficiency and gradient preservation make it a standard choice for hidden layers in deep architectures like CNNs or RNNs.
KKey Concepts to Remember
- Read the scenario before looking for a memorised answer.
- Find the constraint that changes the correct option.
- Eliminate answers that are true in general but not in this case.
TExam Day Tips
- Watch for words such as best, first, most likely and least administrative effort.
- Review why wrong options are wrong, not only why the correct option is correct.
Key takeaway
Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Real-world example
How this comes up in practice
A practitioner preparing for the AI0-001 exam encounters this exact type of scenario on the job. The correct answer here is not the most general option — it is the best answer for the specific constraint described. Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option. Real exam questions reward reading the full scenario before eliminating options, because the constraint defines which answer fits.
Quick reference
OSI Model Reference
| Layer | Name | PDU | Key Protocols / Devices |
|---|---|---|---|
| 7 | Application | Data | HTTP, HTTPS, DNS, SMTP, FTP, SSH |
| 6 | Presentation | Data | TLS / SSL, JPEG, ASCII encoding |
| 5 | Session | Data | NetBIOS, RPC, SIP |
| 4 | Transport | Segment / Datagram | TCP, UDP |
| 3 | Network | Packet | IP, ICMP, OSPF — Routers |
| 2 | Data Link | Frame | Ethernet, Wi-Fi, PPP — Switches, Bridges |
| 1 | Physical | Bits | Cables, NICs, Hubs, Repeaters |
What to study next
Got this wrong? Here's your next step.
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
- →
Machine Learning and Deep Learning — study guide chapter
Learn the concepts, then practise the questions
- →
Machine Learning and Deep Learning practice questions
Targeted practice on this topic area only
- →
All AI0-001 questions
1,000 questions across all exam domains
- →
CompTIA AI+ AI0-001 study guide
Full concept coverage aligned to exam objectives
- →
AI0-001 practice test guide
How to use practice tests most effectively before exam day
Related practice questions
Related AI0-001 practice-question pages
Use these pages to review the topic behind this question. This is how one missed question becomes focused revision.
AI Infrastructure and Technologies practice questions
Practise AI0-001 questions linked to AI Infrastructure and Technologies.
AI Security practice questions
Practise AI0-001 questions linked to AI Security.
AI Concepts and Foundations practice questions
Practise AI0-001 questions linked to AI Concepts and Foundations.
AI Concepts and Techniques practice questions
Practise AI0-001 questions linked to AI Concepts and Techniques.
Machine Learning and Deep Learning practice questions
Practise AI0-001 questions linked to Machine Learning and Deep Learning.
AI Models and Data Engineering practice questions
Practise AI0-001 questions linked to AI Models and Data Engineering.
Implementing AI Solutions practice questions
Practise AI0-001 questions linked to Implementing AI Solutions.
AI Implementation and Operations practice questions
Practise AI0-001 questions linked to AI Implementation and Operations.
AI Security, Ethics and Governance practice questions
Practise AI0-001 questions linked to AI Security, Ethics and Governance.
AI Governance and Ethics practice questions
Practise AI0-001 questions linked to AI Governance and Ethics.
CompTIA A+ hardware practice questions
Practise AI0-001 questions linked to CompTIA A+ hardware.
CompTIA A+ mobile devices practice questions
Practise AI0-001 questions linked to CompTIA A+ mobile devices.
Practice this exam
Start a free AI0-001 practice session
Short sessions build daily habit. Longer sessions build exam-day stamina. Try a timed session to simulate real conditions.
FAQ
Questions learners often ask
What does this AI0-001 question test?
Machine Learning and Deep Learning — This question tests Machine Learning and Deep Learning — Read the scenario before looking for a memorised answer..
What is the correct answer to this question?
The correct answer is: ReLU — ReLU (Rectified Linear Unit) is correct because it outputs zero for negative inputs and a positive linear slope for positive inputs, which avoids the saturation problem of tanh. In deeper networks, tanh gradients can vanish as activations approach ±1, slowing or halting learning. ReLU's non-saturating nature keeps gradients flowing for positive inputs, mitigating the vanishing gradient problem.
What should I do if I get this AI0-001 question wrong?
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
What is the key concept behind this question?
Read the scenario before looking for a memorised answer.
About these practice questions
Courseiva creates original exam-style practice questions with explanations and wrong-answer analysis. It does not publish real exam questions, exam dumps, or protected exam content. Learn why practice questions differ from exam dumps →
Keep practising
More AI0-001 practice questions
- A data science team uses a CI/CD pipeline for ML models. They need to ensure that each model version is traceable back t…
- A company is deploying a large language model for customer support. They want to reduce the number of off-topic or nonse…
- A data scientist fine-tunes a large language model for a legal document summarization task. After fine-tuning, the model…
- A team is designing an AI system for autonomous driving. They need to decide between an end-to-end deep learning approac…
- A team is using Kubeflow to orchestrate ML workflows on Kubernetes. They need to ensure reproducibility, track experimen…
- A healthcare organization deploys an AI system to analyze medical images and detect anomalies. During a routine audit, t…
Last reviewed: Jul 4, 2026
This AI0-001 practice question is part of Courseiva's free CompTIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI0-001 exam.
Question Discussion
Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.
Sign in to join the discussion.