Practice generative model vocabulary for synthetic data: GANs for tabular data, VAE latent space, conditional generation, Wasserstein GAN, and statistical similarity testing.
0 / 10 completed
1 / 10
'The GAN generates realistic tabular data.' How does a GAN (Generative Adversarial Network) work?
A GAN consists of two competing neural networks: the generator (creates synthetic data) and the discriminator (tries to tell real from fake). Training is adversarial — the generator improves to fool the discriminator, the discriminator improves to catch fakes. Eventually the generator produces data that closely matches the real distribution. CTGAN and TVAE are popular GAN variants for tabular data.
2 / 10
'The VAE's latent space captures the data distribution.' What is a Variational Autoencoder (VAE)?
A VAE is a generative model with two parts: an encoder that maps input data to a distribution in a lower-dimensional latent space, and a decoder that samples from that distribution to reconstruct or generate new data. The latent space is continuous and structured, making it useful for generating diverse but realistic synthetic samples — including for tabular data.
3 / 10
What does 'conditional generation (generate samples for class X)' mean?
Conditional generation allows you to control what the generative model produces. Instead of sampling from the full data distribution, you condition on a label or attribute (e.g., class=fraud, age_group=18-25) and generate samples that have those specific characteristics. This is especially useful for augmenting minority classes or generating targeted test cases.
4 / 10
'The Wasserstein GAN training is more stable.' What problem does Wasserstein GAN (WGAN) solve?
Standard GAN training is notoriously unstable — the discriminator can become too good, giving the generator useless gradients (vanishing gradient problem), or the generator collapses to producing only a few samples (mode collapse). WGAN uses the Wasserstein (Earth Mover's) distance as a loss function, which provides a smoother, more informative gradient signal even when the distributions are far apart — leading to more stable training.
5 / 10
'The generated data passes the statistical similarity test.' What does this test verify?
Statistical similarity tests compare the synthetic data to real data across multiple dimensions: marginal distributions of each feature (e.g., Kolmogorov-Smirnov test), pairwise correlations, higher-order statistics, and sometimes downstream ML utility (train on synthetic, test on real). Passing these tests gives confidence that models trained on synthetic data will generalise to real data.
6 / 10
Code Review Comment: 'This data generation is producing extreme outliers. I'm seeing values exceeding 10,000 for the 'transaction_amount' field. It looks like the model isn't respecting any constraints.' What does this comment *most* accurately describe regarding generative models and data?
Outliers in this context refer to values falling significantly outside of expected or defined ranges. The comment highlights a failure of the generative model to respect constraints on specific data fields – a key aspect of controllable generation. The model is not simply generating random numbers; it's producing data that violates established rules, suggesting an issue with its training or configuration.
7 / 10
Slack Message: 'Hey team, we're experimenting with using a diffusion model to augment our customer reviews. The goal is to generate synthetic reviews to improve our training data for sentiment analysis. Anyone have experience with this? Also, how do we ensure the generated reviews sound *authentic*?' What's the primary technical challenge highlighted in this message?
The core challenge is achieving *authenticity* in synthetic data. Diffusion models are powerful but can easily produce outputs that lack nuance and variation if not carefully controlled. The message underscores the need to avoid generating generic or repetitive reviews – a common issue with generative models when they aren't trained effectively for specific tasks.
8 / 10
PR Description: 'Implemented a new API endpoint to allow users to request generated financial transactions. This endpoint accepts parameters like 'amount', 'currency', and 'transaction_type'. The model then generates a synthetic transaction record based on these inputs, returning it as JSON. We've added logging for all requests and responses.' What is the *primary* purpose of the 'transaction_type' parameter?
The 'transaction_type' parameter is crucial for *conditional generation* – directing the model to produce specific types of financial transactions. By controlling this parameter, you effectively guide the generative process towards producing data that aligns with a particular category or operation (e.g., generating only payment transactions versus transfers).
9 / 10
Standup Update: 'I've been working on fine-tuning a generative model to produce synthetic patient data for our research project. The initial results were very noisy – lots of invalid medical codes and unrealistic age ranges. We're now implementing constraints within the generation process to address these issues.' What is the *immediate* problem being addressed in this update?
The update specifically mentions 'noisy' results – indicating a core problem with the generated data: it doesn't conform to expected rules or patterns within the medical domain. Constraints are being added to force the model to adhere to these rules and generate more realistic patient data.
10 / 10
Code Review Comment: 'The generated CSV file contains a significant number of negative values for the 'revenue' column. This is clearly not representative of actual business data. What's going on here?' What potential issue is this comment highlighting concerning generative model output?
This comment points to a fundamental problem: the generative model isn't respecting numerical constraints. 'Revenue,' by its nature, cannot be negative; this violation suggests the model is not properly conditioned to produce realistic and valid data within a specific feature's expected range.
What will I learn from the "Generative Model Vocabulary for Data" exercise?
Practice generative model vocabulary for synthetic data: GANs for tabular data, VAE latent space, conditional generation, Wasserstein GAN, and statistical similarity testing.
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.