Concern 288 · Working draft · editorial review in progress
Serving-vs-training skew
Training-serving skew occurs when a model receives differently prepared inputs in live use than it received during training.
Understanding the concern
Serving means running the model for users. If the training pipeline converts units, fills missing values, or groups categories differently from the live pipeline, the model may interpret inputs incorrectly. The raw information can be unchanged while its representation differs.
This is distinct from drift in the outside world. The mismatch can be created by the system's own code or data processing. Sharing and testing transformations across training and serving helps, along with checking the actual inputs that reach the deployed model.
An example
A model learns temperatures in Celsius but receives numbers prepared as Fahrenheit without the expected conversion.
Illustrative unless identified as a documented case.
Words in plain English
- Model
- The learned part of an AI system. It uses patterns found during training to produce an answer or prediction.
- Training
- The process of using examples to shape how an AI model behaves. Learning patterns from material does not establish that the material is true or fair.
- Drift
- A change over time in data or real-world conditions that can make a system less reliable.
Evidence and its limits
TensorFlow documentation directly describes avoiding training-serving skew through shared transformations and validation.
TensorFlow documentation · Official technical documentation
TensorFlow Transform best practices ↗ (opens in a new tab)Explains shared transformation logic and statistics for training and serving to reduce preprocessing mismatches.
TensorFlow documentation · Official technical documentation
TensorFlow Data Validation: Checking and analyzing your data ↗ (opens in a new tab)Describes data statistics, schema checks, drift and training-serving skew detection; configuration determines what is checked.
Broader frameworks and background
These resources provide context. They do not establish every claim in this concern.
- NIST AI RMF (opens in a new tab)
framework
A question to discuss
Are training and live inputs transformed by the same verified rules?
Keep track of different views, what needs checking, and any next step in your own discussion notes.
Open discussion notes →Principles for the conversation
Before you finish · reflect together
Virtues and principles to consider
Virtues are qualities we practice in how we treat people and make choices. They are not scores for judging someone. Use these ideas to explore what a thoughtful response could look like.
Stewardship
Stewardship means caring for resources and systems that others depend on, including people who will use them later. It looks beyond the benefit of a quick launch.
Plan for maintenance, energy use, data protection, and eventual shutdown. Ask who will look after the tool when the original team moves on and how future users can avoid inheriting hidden costs.
Ask together: What are we responsible for caring for over time?
Prudence
Prudence means using careful judgment before and during action. It balances worthwhile opportunities with the need to avoid harm that could have been anticipated.
Start with a limited trial, set stopping points, and prefer choices that can be undone when uncertainty is high. Review what happens rather than assuming the original plan remains sound.
Ask together: What is a careful next step, and when should we stop or change course?
Cooperation
Cooperation means working with others toward a shared purpose while respecting their contribution. It is not simply asking people to accept a decision already made.
Share information, roles, and credit. Include the people who do the daily work when choosing or testing a tool. Make disagreements discussable so that cooperation does not become pressure to agree.
Ask together: What can we do together that none of us can do well alone?
Editorial ratings
These are editorial judgments from the register, not measured probabilities or predictions about your situation.
About the register and scoring →More ways to explore
Search for more writing
For more specific or recent material, open a search below or ask for a web summary. AI-generated summaries can contain mistakes; open the original articles to check them.