Construct · 5 fields · 6 citations

Where does drift collide?

One name. Different field. Different construct. One map.

● jingle · one word, five constructs Nearest construct by meaninggenetic driftcompare the meanings
How to read this

Every construct on the map is compared with every other by the meaning of its definitions, and the closest one is surfaced here. The nearest construct to drift is genetic drift.

Hold this as a prompt, not a verdict. Two constructs sitting close in meaning is a reason to read them side by side and decide for yourself whether they are one idea under two names. It is not proof that they are. Compare the meanings in the Collider.

drift5 CONSTRUCTSGenetic driftWRIGHT · ALLELESConcept driftGAMA · MODEL DECAYContinental driftWEGENER · CONTINENTSInstrument driftJCGM · SENSOR CREEPDriftDEKKER · SAFETY SCIENCE

Spokes are fields; dashed arcs are shared lineage. Tap a field for its definition, unit, and measures.

2026

Drag to watch the word get claimed. Each field appears at the publication date of the earliest source cited for it. Tap a year to see which.

Population genetics

Random change in allele frequencies

Random change in the frequency of an allele from one generation to the next, arising because a finite population passes on only a sample of its genes. Wright set the idea out in his 1931 Evolution in Mendelian Populations, where it works alongside selection, mutation, and migration. The referent is an allele frequency in a population; the change has no direction and no adaptive purpose, and its strength scales inversely with population size, so that in small populations frequencies wander fast and variation is lost. It differs from selection, which pushes frequencies in a direction: genetic drift is pure sampling error, a change caused by chance rather than by any advantage.[1]

Unit
The frequency of an allele in a finite population
Recovery means
Not applicable
Measured by
Change in allele frequencies across generations; the per-generation variance is about 1 over twice the effective population size

Machine learning

The input-to-target relation shifting over time

The change over time in the statistical relationship between a model's inputs and its target, which degrades a model that was fit to the old relationship. Gama, Žliobaitė, Bifet, Pechenizkiy, and Bouchachia surveyed it in 2014 in A survey on concept drift adaptation, building on Widmer and Kubat, who in 1996 studied learning under hidden contexts that change. The referent is the joint distribution of inputs and target, or the conditional distribution of the target given the inputs, that a deployed model depends on. Nothing is copied across generations; what moves is the world the model was trained on. Unlike genetic drift, the shift is often driven by an outside cause and is not undirected sampling error.[2][3]

Unit
The conditional relation of target to inputs that a model relies on
Recovery means
Relevant: an adaptive learner detects the change and updates or retrains the model to recover accuracy
Measured by
Falling predictive accuracy on incoming data, and statistical tests comparing the input or error distributions across time windows

Geology

Continents moving over the Earth

Wegener's hypothesis that the continents are not fixed but move across the surface of the Earth, once joined in a single landmass and since parted. He proposed it in 1912 and developed it in The Origin of Continents and Oceans. The referent is a continent's position on the globe, changing over millions of years. The mechanism Wegener offered was wrong, and the idea was later subsumed by plate tectonics, but the drift itself, the motion of the continents, is real and is now measured by geodesy in centimetres per year. This is literal movement of matter across space, not a change in a frequency or a probability distribution.[4]

Unit
A continent's position on the Earth's surface
Recovery means
Not applicable
Measured by
Reconstructed past positions from fossils, rock and glacial records; today, plate velocities from satellite geodesy in centimetres per year

Metrology

A sensor's baseline creeping under constant input

A slow, unwanted change in a measuring instrument's output over time while the quantity being measured is held constant. The International Vocabulary of Metrology defines instrumental drift as a continuous or incremental change in indication that comes from changes in the instrument itself, in its temperature or its ageing components, rather than from any change in the measurand. The referent is a sensor's reading under a fixed reference input; the change is a defect to be corrected, and recalibration against a standard removes it. Here drift is a property of an instrument, not of a population, a model, or a planet, and it is unwanted by definition.[5]

Unit
A sensor's indicated output under a constant reference input
Recovery means
Relevant: recalibration against a known reference removes accumulated drift
Measured by
The change in indicated output over time while a stable reference input is applied, expressed per hour, per day, or per year

Safety science

drift into failure, the slow erosion of safety margins in complex systems through decisions that each seem reasonable at the time.

In safety science, drift refers to drift into failure, the slow erosion of safety margins in complex systems through decisions that each seem reasonable at the time. This meaning comes from Dekker (2010) in Drift into failure.[6]

Unit
Recovery means
Measured by

Next

Run it through the Collider, or take the citations with you.

See how drift sits against neighbouring constructs in the Collider, or export every reference on this page in one click.

Open in ColliderExport BibTeXReport this collision

Nearby collisions

selection4 fields · 5 citationsjingle · one word, four constructsmutation4 fields · 4 citationsjingle · one word, four constructsnoise4 fields · 5 citationsjingle · one word, four constructs

References

  1. Wright, S. (1931). Evolution in Mendelian populations. Genetics, 16(2), 97–159. https://doi.org/10.1093/genetics/16.2.97
  2. Gama, J., Žliobaitė, I., Bifet, A., Pechenizkiy, M., & Bouchachia, A. (2014). A survey on concept drift adaptation. ACM Computing Surveys, 46(4), Article 44. https://doi.org/10.1145/2523813
  3. Widmer, G., & Kubat, M. (1996). Learning in the presence of concept drift and hidden contexts. Machine Learning, 23(1), 69–101. https://doi.org/10.1023/a:1018046501280
  4. Wegener, A. (1966). The origin of continents and oceans (J. Biram, Trans.). Dover Publications. (Original work published 1929)
  5. Joint Committee for Guides in Metrology. (2012). International vocabulary of metrology: Basic and general concepts and associated terms (VIM) (3rd ed., JCGM 200:2012). JCGM.
  6. Dekker, S. (2010). Drift into failure. Taylor & Francis Group. ISBN 9781351942904

Cited on this page?

Confirm your construct, or correct us.

If your work is cited behind one of these entries, you know better than we do whether we captured the meaning right. Confirm it, suggest an edit, or dispute it, and your name goes on the entry as an author who has checked it, linked to your ORCID.

Claim your construct