Model bias
Systematic, unfair error in a model's outputs, usually inherited from skewed training data or design choices.
At a glance
Key signals
0% cross fields · reaches 4 more
- Machine Learning
- Ethics
- Statistics
- Explanation
- Examples
- Misconception
- Sourced relations
- Attribution
Dependencies
What this concept builds on and what it makes possible — derived from the atlas’s dependency, causal and structural relations, not from every related edge.
Foundations · builds on
Model biasis derived fromTraining dataStrongly supported
A model's bias is largely inherited from skewed training data.
Mechanism: Skew in the examples becomes skew in the model: it learns the imbalance as if it were the truth.
Structural role & consequence
Interpreted from the current atlas graph — what the connections mean, not just how many there are.
Currently dark in the atlas: no key date stored · 2 of 2 of its relations lack claim-level evidence.
atlas representation · Describes the current Thinking OS representation, not the state of the world.
Builds on 1 foundation (requires / depends-on / derived-from / emerges-from).
structural · Structural graph analysis — not a claim of importance, causation or history.
Exercises 2 annotated mental models — a concept that connects several thinking patterns.
curated · Curated annotations, not a derived measure.
Structural neighbourhood: 2 → 14 → 75 concepts reachable within 3 hops.
structural · Structural reach — being reachable is not the same as being understood.
cross-field
2 within-field, 0 cross-field
Strengths & constraints
Constraints
- Evidence coverage currently thin in the atlas — few of its relationships carry claim-level evidence. atlas representation
- No dated history stored — the atlas records no key date for this concept. atlas representation
Conditions
- Its dependency reading rests on 1 foundation relation. structural
- Read structurally — most of its relationships carry no external evidence yet, so claims here are graph-derived. structural
Dependency radial
What this concept builds on (left) and what it makes possible (right) — derived from dependency and causal relations.
Seen through each discipline
How this concept sits in each of its fields — derived from its real connections in the graph, not asserted.
Through this lens it connects to Training data.
Through this lens it connects to Risk and Training data.
Check yourself
A quick check against a common misconception. Nothing is scored — picking the tempting-but-wrong answer just flags an idea worth revisiting.
Which statement is correct?
Because a computer made the decision, it must be neutral and objective.
A model reflects its data and design. It can be systematically unfair while looking perfectly objective.
Look for: Learner treats a model's output as objective because 'a computer decided', ignoring the data and design behind it.
Where you'll meet it
Journeys that walk you through this idea. You may recognise it from more than one.
Related ideas to explore
Concepts that look related but are not yet connected here — candidates for a connection to reason about, not established links.
This idea also appears in…
The same structure shows up in other disciplines. These are real recurrences drawn from the graph — a starting point for asking “what carries over, and what changes?”
Probability23 disciplines · 21 concepts
Cause and effect31 disciplines · 31 concepts
Concepts
shares a mental model · shares a mental model
shares a mental model · shares a mental model
shares a mental model · shares a mental model
shares a mental model
shares a mental model
shares a mental model
A model learns from whatever it is shown. If the examples over-represent some groups or situations, the model quietly carries that skew into its decisions — and can amplify it at scale. This kind of bias is not an opinion; it is a measurable, systematic error with real consequences.
Mental models at work here
Probability
A way to reason about uncertainty by assigning each possible outcome a share of the whole, between impossible (0) and certain (1).
Instead of 'will it happen?', ask 'how often would it happen if this repeated many times?'
Cause and effect
One thing genuinely bringing about another — as opposed to two things merely moving together. Establishing it needs a mechanism and controls, not just a pattern.
Ask: if I changed the cause, would the effect change? And what else might explain the link?
- Connects 2 other ideas across 4 disciplines.
- A cross-disciplinary bridge — its connections reach into 4 other fields.
- Most of its connections are of the “Cause & effect” kind.
- It exercises 2 reusable thinking patterns.
- It features in 1 learning journey.
Derived from the graph’s real structure — observations, not a score.
Sources
- On the bias–robustness in the location model I (1989) verified