Practice privacy-preserving machine learning vocabulary: federated learning, differential privacy, secure multi-party computation, synthetic training data, and epsilon-DP guarantees.
0 / 10 completed
1 / 10
'Federated learning avoids sharing raw data.' How does federated learning achieve this?
In federated learning, the model is sent to where the data lives (e.g., mobile devices, hospital servers) and trained locally. Only the model updates (gradients or weights) are sent back to a central aggregator — never the raw data. This allows collaborative model training across sensitive datasets that can't be centralised due to privacy regulations.
2 / 10
'Differential privacy adds calibrated noise.' What is the purpose of this noise?
Differential privacy (DP) provides a mathematical privacy guarantee: the statistical output of a query or model is nearly indistinguishable whether or not any individual's data is included. This is achieved by adding carefully calibrated random noise (Laplace or Gaussian) to the computation. The epsilon parameter controls the privacy-utility trade-off — lower epsilon means stronger privacy but more noise.
3 / 10
What is 'secure multi-party computation' (SMPC) in the context of privacy-preserving ML?
Secure multi-party computation (SMPC) allows multiple parties to collaborate on a computation — such as training a model or computing statistics — where each party holds private data, and no party learns anything about the others' data beyond the final result. It uses cryptographic protocols (secret sharing, garbled circuits) to achieve this.
4 / 10
'The model is trained on synthetic data only.' What privacy benefit does this provide?
Training on synthetic data instead of real data can provide privacy benefits by ensuring the model never directly processes sensitive personal information. However, this only works if the synthetic data generator itself doesn't memorise or leak sensitive patterns from the real data — poorly generated synthetic data can still carry privacy risks through statistical disclosure.
5 / 10
'We achieve epsilon-differential privacy.' What does the epsilon value represent?
Epsilon (ε) is the differential privacy budget — it quantifies the maximum privacy loss guaranteed. A lower epsilon provides stronger privacy (the output changes very little even if one person's data is added or removed) but requires more noise, which reduces model utility. Epsilon values in practice range from 0.1 (very strong privacy) to 10+ (weaker, more utility-preserving).
6 / 10
Reviewer: 'I noticed you're using a local copy of the user data for this model training. Are you considering any techniques to minimize its exposure?'
What is the primary concern driving this reviewer's question regarding the use of a local copy of user data?
The reviewer is highlighting a fundamental risk: exposing raw user data, even temporarily, can lead to privacy breaches and potential misuse. The core concern isn't about performance or simplicity, but about the security and ethical implications of storing sensitive information where it's easily accessible. This aligns with the broader goal of privacy-preserving ML.
7 / 10
Alex (Data Scientist): 'Just ran some initial tests on our federated learning model – it's showing promising results but the variance between client updates is pretty high. We're aiming for low noise contribution.'
What does Alex likely mean by 'low noise contribution' in this context?
In federated learning, 'noise contribution' refers specifically to the amount of calibrated noise added to client updates before aggregation. Reducing this noise is crucial for achieving differential privacy – less noise means better protection for individual user data while still allowing the model to learn effectively. It's not about accuracy or convergence speed.
8 / 10
PR Title: 'Implemented SMPC for key feature extraction.'
Which of the following best describes the role of 'Secure Multi-Party Computation' (SMPC) in this PR's context?
SMPC allows multiple parties to jointly compute a function (like extracting features) on their private datasets without ever revealing those datasets to each other. This is key to privacy-preserving ML because it avoids direct data sharing and aggregation, fundamentally different from simply encrypting the model or datasets. It's about computation itself.
9 / 10
API Response (from a synthetic data generator): `{"synthetic_data_version": "v3", "data_size": 100000, "noise_level": 0.05}`
What does the 'noise_level' parameter in this API response primarily indicate when generating synthetic data for ML training?
The 'noise_level' parameter directly controls the amount of artificial variation introduced into the synthetic data. This is essential for creating a dataset that's sufficiently different from the real data to prevent model memorization and improve privacy—effectively mimicking real-world variations with calibrated uncertainty.
10 / 10
Ben (ML Engineer): 'We're using a generative model to create synthetic training data. We've defined epsilon equal to 0.1 for differential privacy.'
What does the value 'epsilon = 0.1' represent in Ben's statement concerning differential privacy?
In differential privacy, 'epsilon' (often denoted as ε) is a crucial parameter that quantifies the level of privacy protection. A smaller epsilon value means stronger privacy guarantees because it introduces more noise to mask individual data points – effectively limiting the potential for an attacker to infer information about specific users based on the model's output.
What will I learn from the "Privacy-Preserving ML Vocabulary" exercise?
Practice privacy-preserving machine learning vocabulary: federated learning, differential privacy, secure multi-party computation, synthetic training data, and epsilon-DP guarantees.
Is this exercise free to use?
Yes. Every exercise on CoderSlingo, including this one, is free to use with no account, sign-up, or paywall required.
How many questions are in this exercise?
This set contains 10 multiple-choice questions, each with a detailed explanation shown after you answer.
Do I need to create an account to track my progress?
No account is required. Your progress bar and score reset each time you reload the page, but you can retry the exercise as many times as you like.
Who is this Synthetic Data Vocabulary exercise for?
This exercise is built for IT professionals and non-native English speakers who need to read, write, and discuss synthetic data vocabulary topics confidently at work.
What happens if I answer a question incorrectly?
You will see the correct answer highlighted along with a detailed explanation of why it is correct -- so every wrong answer becomes a learning moment, not just a lost point.
Can I retry this exercise?
Yes -- click "Try again" on the results screen at any time to reset your score and go through all the questions again.
How long does this exercise take to complete?
Most learners finish all 10 questions in under 10 minutes, since each question is answered by clicking a single option.
Where can I find more Synthetic Data Vocabulary exercises?
See the full Synthetic Data Vocabulary exercises hub for more vocabulary drills on this topic.
Is this exercise mobile-friendly?
Yes -- the exercise works on any device with a modern browser, including phones and tablets, with no app download required.