
01 / Clustering · Unsupervised learning
Starlight Mixer
Find patterns in similarity
Tune the resonance, compare six signals, and see similar signals form clusters.
Data can reveal structure even without answer labels.
What’s inside
Find a pattern, shape a story, or check a rescue route. These are the learning missions released and available to play today.
All released missions are currently open. When purchases open, Missions 01–05 will remain free. Pricing

01 / Clustering · Unsupervised learning
Tune the resonance, compare six signals, and see similar signals form clusters.
Data can reveal structure even without answer labels.

02 / Examples & labels · Supervised learning
A flight reaches the wrong home. Inspect Dogo’s examples, correct a label, then test with a new traveler.
Training examples and their labels influence predictions for new cases.

03 / Reward feedback · Reinforcement learning
Watch Dogo use a gadget, decide whether to reward the result, and observe the next choice.
Rewards after an action can influence future strategies.

04 / Outcomes & strategies · Reinforcement learning
Load a route chip, predict and observe Dogo’s action, compare choices before and after a failure, then test similar new conditions.
Past outcomes can influence later choices in similar situations.

05 / Combining three ways of learning
Help a traveler get home by comparing signals, using examples, and adjusting actions after feedback.
Different learning methods address different problems within one task.

06 / Data coverage · Generalization
When night falls at the harbor, inspect boarding decisions, add useful examples, and test with new travelers.
Success in familiar conditions does not guarantee reliability in a new setting.

07 / Context · Step-by-step generation
Change a clue already written in the theater, compare the next candidates, and watch a story continue.
What is already in the context influences what comes next.

08 / Generated errors · Fact-checking
Check each suggested route segment against records and field evidence, then build a supported rescue route.
A fluent, confident suggestion is not necessarily supported by facts.

09 / Reward design · Specification gaming
Try cleanup rules yourself, compare points with correctly placed objects, then revise the reward and test a new room setup.
A high score does not necessarily achieve the outcome people intended.
This is the complete list of missions available to play now. It is updated when a mission is released.